Professional Documents
Culture Documents
Author: Dr. Susanne Friese (QUARC Consulting) Realization/Programming: Dr. Thomas G. Ringmayr (www.hypertexxt.com)
Copying or duplicating this manual or any part thereof is a violation of applicable law. No part of this manual may be reproduced or transmitted in any form or by any means, electronic or mechanical, including, but not limited to, photocopying, without written permission from ATLAS.ti GmbH. Trademarks: ATLAS.ti is a registered trademark of ATLAS.ti Scientific Software Development GmbH. Adobe Acrobat is a trademark of Adobe Systems Incorporated; Microsoft, Windows, Excel, and other Microsoft products referenced herein are either trademarks of Microsoft Corporation in the United States and/or in other countries. Google Earth is a trademark of Google, Inc. All other product names and any registered and unregistered trademarks mentioned in this document are used for identification purposes only and remain the exclusive property of their respective owners.
Contents
Introduction..............................................................................................................................9 Main Concepts.......................................................................................................................10
The Hermeneutic Unit (HU)............................................................................................................................................10 Primary Documents (PD or PDoc)...................................................................................................................................11 Primary Document Families (Data Attributes)..................................................................................................................12 Quotations.....................................................................................................................................................................12
Quotations as Layers.......................................................................................................................................................................13 Quotation references.......................................................................................................................................................................13 Types of Quotations........................................................................................................................................................................14 Text Quotations..........................................................................................................................................................................14 PDF Quotations..........................................................................................................................................................................15 Audio and Video Quotations......................................................................................................................................................15 GoogleEarth Quotations.............................................................................................................................................................16 Coding Objectives...........................................................................................................................................................................16 Code Reference...............................................................................................................................................................................17 Text Search Tool..............................................................................................................................................................................18 Word Cruncher................................................................................................................................................................................18 Auto Coding Tool............................................................................................................................................................................18
Codes.............................................................................................................................................................................16
Analysis .........................................................................................................................................................................20
Team Tools.....................................................................................................................................................................22
Export.............................................................................................................................................................................23
Miscellaneous Tools........................................................................................................................................................24
Working with Internal Documents (for smaller projects containing only text files) ..........................................................27 The Process.....................................................................................................................................................................28 General Steps when Working with ATLAS.ti...................................................................................................................29
Data-Level Work.............................................................................................................................................................................28 Conceptual Level Work....................................................................................................................................................................29
Project Management..............................................................................................................31
Project Setup..................................................................................................................................................................31 Saving the Project and Controlling Project Setup.............................................................................................................32 Working in Teams...........................................................................................................................................................33
Project administration......................................................................................................................................................................34 Creating user accounts.....................................................................................................................................................................34 Logging in.......................................................................................................................................................................................34 Backup options for the HU..............................................................................................................................................................36 Auto recovery file............................................................................................................................................................................36 Backup of. file..........................................................................................................................................................................36 Customizing HU backup options......................................................................................................................................................37 When to create a copy bundle file...................................................................................................................................................37 Consider file size when creating copy bundle files...........................................................................................................................38 Potential problems when creating a bundle file................................................................................................................................39 Installing an acb (atlas copy bundle) file..........................................................................................................................................39 Possible conflicts..............................................................................................................................................................................40 Changing absolute path references to the optimal HUPATH...........................................................................................................42 Exchanging doc or docx files with rtf files (changing paths).............................................................................................................42 Required data file format.................................................................................................................................................................43 What is changed?............................................................................................................................................................................43 Mind the following rules!.................................................................................................................................................................43 Rule No. 1 Always edit within ATLAS.ti!.......................................................................................................................................43 Rule No. 2: Never delete or modify the auxiliary log file!.................................................................................................................44 How to edit.....................................................................................................................................................................................45 To exit edit mode:............................................................................................................................................................................45 Editing primary documents in team situations .................................................................................................................................45 Editing in teams using a server-based project setup.........................................................................................................................46 Editing in teams when working at different locations.......................................................................................................................46 If the project administrator only is allowed to edit...........................................................................................................................46 Best practice rules for editing primary documents............................................................................................................................47
Modifying An Existing Project Setup...............................................................................................................................42 What You Should Know About Editing...........................................................................................................................43
Step 1: Create an audio or video quotation ...............................................................................................................................55 Step 2: Code an audio or video quotation..................................................................................................................................57 Reviewing audio and video quotations.......................................................................................................................................57 Unlinking codes..........................................................................................................................................................................57 Modifying the length of a video quotation.................................................................................................................................57 Resetting the filter......................................................................................................................................................................57 Describing video quotations to improve outputs..............................................................................................................................58 Renaming video quotations and creating text output.................................................................................................................58 Changing between time and frame references............................................................................................................................58 Making use of quotation comments...........................................................................................................................................58
Family Life..............................................................................................................................59
Purpose of Creating and Working with Families..............................................................................................................................59 Super Families..................................................................................................................................................................................60
Common Procedures......................................................................................................................................................60
The Family Manager........................................................................................................................................................................60 Components of the Family Manager...........................................................................................................................................61 To Open a Family Manager........................................................................................................................................................61 Creating Families.............................................................................................................................................................................62 Adding Members ............................................................................................................................................................................62 Removing Members.........................................................................................................................................................................63 Using Families as Filters....................................................................................................................................................................63 Access the Filter Options via the Main Menu..............................................................................................................................63 Filter Items using the Family Manager........................................................................................................................................63 Removing Families...........................................................................................................................................................................65
Visualization...........................................................................................................................99
Tools for Visualization and Theory Building.....................................................................................................................................99
Network Views...............................................................................................................................................................99
Nodes and Links............................................................................................................................................................................100 Directed and Non-Directed Links..............................................................................................................................................100 Strong and Weak Links.............................................................................................................................................................102 Network vs. Network View.......................................................................................................................................................102 Node Types...............................................................................................................................................................................102 Relations........................................................................................................................................................................................102 Link vs. Relation........................................................................................................................................................................103 The Role of Relations................................................................................................................................................................104 The Network Editor ..................................................................................................................................................................104 Basic Network View Procedures.....................................................................................................................................................105 Creating Network Views...........................................................................................................................................................105 To create a new Network View............................................................................................................................................105 To open a Network View on an Object................................................................................................................................106 Adding Nodes to a Network View............................................................................................................................................106 Via Drag & Drop..................................................................................................................................................................106 Via the Import Nodes dialog.................................................................................................................................................107 Linking Nodes...........................................................................................................................................................................108 To link two nodes using the toolbar icon..............................................................................................................................108 Cutting Links.............................................................................................................................................................................109 Modifying Links........................................................................................................................................................................110 To remove nodes from a Network View...............................................................................................................................110 To delete nodes from a Network View.................................................................................................................................110 Simple analytic functions...........................................................................................................................................................111 Import Node Neighbors........................................................................................................................................................111 Import Co-occurring Codes..................................................................................................................................................111 Creating Output............................................................................................................................................................................111 Printing Networks.....................................................................................................................................................................112 Network Views for other Applications.......................................................................................................................................112 Copy to Clipboard................................................................................................................................................................112 Save Network View as Graphic File......................................................................................................................................112 Representing the Rhetoric of Text.............................................................................................................................................114 Benefits of Hypertext.....................................................................................................................................................................115 What codes cannot do..............................................................................................................................................................115 Graphical Hyperlink Maps.........................................................................................................................................................116 General Procedures...................................................................................................................................................................117 Star or Chain connections.....................................................................................................................................................117 Hyperlinks in the Quotation Manager.......................................................................................................................................118 Hyperlinks in the Margin Area..................................................................................................................................................118 Creating Hyperlinks.......................................................................................................................................................................119 Creating Hyperlinks using the context menu............................................................................................................................119 To create a chain:.................................................................................................................................................................119 To create a star:....................................................................................................................................................................120 Creating Hyperlinks in the Quotation Manager........................................................................................................................120 Defining New Hyperlink Relations............................................................................................................................................121 Traversing Hyperlinks.....................................................................................................................................................................121 To traverse hypertext links using the margin area................................................................................................................121 The "In-Place" Method for audio- and video hyperlinks..........................................................................................................122 To display hyperlinked neighbors "in-place"........................................................................................................................122
Hypertext.....................................................................................................................................................................114
Generating Output...............................................................................................................124
Overview......................................................................................................................................................................124 General Procedure........................................................................................................................................................125
Output Destinations......................................................................................................................................................................125 Report Layout................................................................................................................................................................................125 Common Header......................................................................................................................................................................125 Sorts & Filters............................................................................................................................................................................126 Restricting Output to Selected Objects.....................................................................................................................................126 Working with Style Sheets.............................................................................................................................................................127 Example: Code book ...............................................................................................................................................................128
Appendix..............................................................................................................................135
Useful Resources...........................................................................................................................................................135
The ATLAS.ti Website....................................................................................................................................................................135 Quick Tour....................................................................................................................................................................................135 Full Manual and How To Documents........................................................................................................................................135 Video tutorials...............................................................................................................................................................................136 Library...........................................................................................................................................................................................136 Knowledge Base............................................................................................................................................................................136 Social Media..................................................................................................................................................................................136 YouTube...................................................................................................................................................................................136 Facebook..................................................................................................................................................................................136 Twitter......................................................................................................................................................................................136 Newsroom.....................................................................................................................................................................................137 The Help Menu..............................................................................................................................................................................137 Getting Support.............................................................................................................................................................................137 Frequently Asked Questions..........................................................................................................................................................138 Forum............................................................................................................................................................................................138 Mailing List Archive.......................................................................................................................................................................138 Help with Data Management Problems.........................................................................................................................................139 Documents cannot be accessed................................................................................................................................................139 Embedded Objects Cannot be Activated..................................................................................................................................139 Cannot Enter Edit Mode for Primary Documents......................................................................................................................139 Reporting Bugs.........................................................................................................................................................................139 To run the software as administrator (for VISTA and Windows 7 users)........................................................................................141 Live Update Settings......................................................................................................................................................................141 Authorship.....................................................................................................................................................................................143 Auto Coding Tool..........................................................................................................................................................................143 Backup...........................................................................................................................................................................................143 Codes............................................................................................................................................................................................143 Coding Analyzer............................................................................................................................................................................143 Comments.....................................................................................................................................................................................143 Cooccurence Explorer....................................................................................................................................................................144 Copy Bundle..................................................................................................................................................................................144 Families..........................................................................................................................................................................................144 Family Manager.............................................................................................................................................................................145 Hermeneutic Unit..........................................................................................................................................................................145 Hermeneutic Unit Editor................................................................................................................................................................145
Troubleshooting............................................................................................................................................................139
Hermeneutic Unit Merger..............................................................................................................................................................145 HTML Generator...........................................................................................................................................................................145 Link Managers...............................................................................................................................................................................145 Margin Area..................................................................................................................................................................................146 Memos..........................................................................................................................................................................................146 Network Editor..............................................................................................................................................................................146 Network Views..............................................................................................................................................................................146 Network View Manager................................................................................................................................................................146 Nodes, Links & Relations...............................................................................................................................................................146 Object Crawler..............................................................................................................................................................................147 Object Explorer..............................................................................................................................................................................147 Object Managers...........................................................................................................................................................................147 Primary Documents.......................................................................................................................................................................147 Primary Document Families (Data Attributes)................................................................................................................................147 Project Transfer..............................................................................................................................................................................148 Query Tool....................................................................................................................................................................................148 Quotations....................................................................................................................................................................................148 Relation Editor...............................................................................................................................................................................149 Super Code....................................................................................................................................................................................149 Text Editor.....................................................................................................................................................................................149 Text Search Tool............................................................................................................................................................................149 Variables........................................................................................................................................................................................149 Word Cruncher..............................................................................................................................................................................149 XML Converter..............................................................................................................................................................................150 XML Generator..............................................................................................................................................................................150 User Administration.......................................................................................................................................................................150
INTRODUCTION
Introduction
This document provides a quick overview of the basic concepts underlying the work with ATLAS.ti. It is designed to provide the big picture, and in doing so aims at helping you to quickly get your bearings when first starting out your work with the program. Proper understanding of the central concepts and functions is key to avoiding some of the most common pitfalls. We have therefore compiled those passages from the full manual that (a) introduce the major concepts and their corresponding functions (such as codes, quotations, memos, families, etc.), (b) provide practical tips in how to approach your projects (e.g. data and project management), and (c) have added a comprehensive glossary that provides succinct overviews of all the central terms, concepts, and functions that you encounter in your work with ATLAS.ti. The full manual is available from http://manual.atlasti.com We also recommend the following essential resources for more detailed information, additional help, and for practical guidance in getting started: Quick Tour: Free Webinars: Video Tutorials: Knowledge Base: Training Center: Workshops: Support Center: http://quicktour.atlasti.com http://webinars.atlasti.com http://youtube.atlasti.com http://kb.atlasti.com http://training.atlasti.com http://workshops.atlasti.com http://support.atlasti.com
MAIN CONCEPTS
10
Main Concepts
MAIN CONCEPTS
11
An HU can become a highly connected entity, a dense web of primary data, associated memos and codes, and interrelations between the codes and the data. To find your way through this web, ATLAS.ti provides powerful browsing, retrieval and editing tools.
MAIN CONCEPTS
12
Quotations
A quotation is a segment from a PD that is interesting or important to the user. In textual documents, a quotation is an arbitrary sequence of characters ranging from a single character, a word, a sentence, or a paragraph up to the entire data file. Free quotations resemble passages "scribbled" in the margin of a book. Usually quotations are created by the researcher. However, if repetitive words or phrases are found in the text, the Auto-Coding feature can be used to automatically segment these quotations and assign a code to them. When a quotation is created, ATLAS.ti automatically assigns an identifier to it. This identifier is built from the index of the primary text to which it belongs and the first 30 letters (note that a different length can be set via Preferences) of the text segment, e.g., "1:21 Therefore a more efficient fil....". The identifier is displayed in list windows and printouts. For graphic, audio, and video segments, the original file name of the PD is chosen as an identifier. Although the creation of quotations is almost always part of a broader task like coding or writing memos, "free" quotations can be created that indicate interesting parts in the primary data for which a meaningful classification has not yet been found. See The Textual Level Basic Functions in the full manual.
MAIN CONCEPTS
13
Quotations as Layers
Quotations are stored inside the HU, independent of the document to which they belong. Quotations need to be regarded as a transparent layer on top of a document. Quotations are stored independently of the PD inside the HU. Technically speaking, a quotation consists of the identifier (a number) and a pair of coordinates that specify the beginning and end of the quotation. The content of a PD file (the data source) is therefore not altered by the creation, deletion, or modification of quotations. Read more on quotations in our library: http://downloads.atlasti.com/library/Maietta_2009-05_6.pdf
Quotation references
The bars in the margin mark the length of the quotation graphically. The entry in the quotation field shows a textual reference for the quotation. It consists of the following elements: ID, name, start and end position. ID: The quotation ID is composed of the number of its P-Doc and a second number indicating when the quotation was created. The ID 8:1 means that the quotation is from P-Doc 8 and is the first one that was created in this document. The reason for the chronological numbering is to do with the fact that you will not necessarily code a document from the first line to the last. You will jump between passages and modify or delete some quotations during the coding process. A linear numeration would have to be updated with every single quotation that is inserted, which would take up unnecessary computation capacity. Sorting the quotations by their start position, for example, offers a clear linear view of your quotes. Recently an option was added to change the chronological numbering to a sequential order if need be (QUOTATIONS / MISCELLANEOUS / RENUMBER ALL TO DOCFLOW). This can for example be useful when coding open ended questions from survey data and you want to keep the cases in synch with the cases in the SPSS file. Name: The name shows the first 30 characters of a textual quotation. This is the default setting which can be changed under EXTRAS / PREFERENCES / GENERAL PREFERENCES. Quotations based on image, audio or video files show the file name. The name of a quotation can be renamed. This is a useful option for image, audio and video quotations, as we will see below. Start and end positions: The figures in brackets after the quotation name show the location (start and end position) in the document. For textual quotations,
MAIN CONCEPTS
14
the reference given is to the paragraph numbers within which the coded segment occurs. (9:9) thus means that the quotation starts and ends in paragraph 9. For image files, the rectangular area marked as the quotation is referenced. Audio quotes use a time reference, and for video quotations you can choose between time or frame numbers. References for PDF quotations consist of: page number and number of characters on the page for start and end positions. In case the document contains columns, the column number is provided as well. Video tutorial: http://www.atlasti.com/video/article/creatingquotations.html
Types of Quotations
In accordance with the six different types of PDs, there are also six different types of quotations:
Text Quotations
A textual quotation originates from an arbitrary sequence of selected characters. Textual quotations represent (for the computer) a sequence of characters ("strings") and can be of arbitrary size. Sentences, speech turns, or paragraphs are often the basis for the length of textual quotations. Only text offers enough "syntactical clues" to allow for searches for the occurrence of specific evidence that may support a concept. Text also offers the option for automatic segmentation as used by the Auto-Coding procedure (see the section on The Auto-Coding Tool in the full manual). Graphic Quotations
The creation, activation, and display of graphical quotations has similarities with, but also differs from, their textual counterparts. A graphical quotation is a rectangular region inside a graphical PD. From its data structure, it is identical to textual quotations since their main attributes are also the PD identifier and two coordinates that mark the beginning and end, defining a rectangle through its upper left and lower right corner.
MAIN CONCEPTS
15
Handling graphical quotations is analogous to marking text passages in a textual document. See reating Graphical on page 158.
PDF Quotations
PDF quotations can be of a textual or of a graphical nature. The quotation references for textual quotations indicate the page number and the start and end position on the basis of character counts For example: (31:1537-31:1745) means that this quotation is from page 31, starting at character 1537 and ending at character 1745. The reference for coded images indicates the position of the quotation within the PDF file like: (@422-@618). Read more about using PDF documents in ATLAS.ti in our library: http://downloads.atlasti.com/library/Friese_2009-05_18.pdf
Audio and video quotations can be as short as a few milliseconds. The length of a quotation can be selected on a time line. Segment starting points and length are displayed in the following formats: milliseconds / HH:MM:SS:ms / frames (for videos)
MAIN CONCEPTS
16
GoogleEarth Quotations
When creating a GE quotation, you see the ATLAS.ti icon on the GE map. It is linked to one location and thus is different from all other types of quotations that present a range. The quotation ID shows either the name of the document or the feature's name (if applicable). In addition the geographic reference of the marked location is provided. See our library for an example on how to work with GoogleEarth data: http://downloads.atlasti.com/library/Cisneros_2009-03_17.pdf
Codes
The term code is used in many different ways. First we would like to define what that term means in qualitative research and then in ATLAS.ti.
Coding Objectives
From a methodological standpoint, codes serve a variety of purposes. They capture meaning in the data. They also serve as handles for specific occurrences in the data that cannot be found by simple text-based search techniques. Codes are used as classification devices at different levels of abstraction in order to create sets of related information units for the purpose of comparison (e.g., a concept like "Coping Strategy"). Keep code names brief and succinct. Use the comment pane for longer elaborations. From a "low level" tool perspective, codes are typically short pieces of text referencing other pieces of text, graphical, audio, or video data. Their purpose is to classify an often large number of textual or other data units.
MAIN CONCEPTS
17
In the realm of information retrieval systems, the terms "index," "indexing," or "keyword" are often used for what we call "code" or "coding". The length of a code should be restricted and should not be too verbose. If textual annotations are what you want, you should use quotation comments instead. The technical aspects of coding are described under Textual Level Basic Functions, p. 116 ff in the full manual. You find information on how to structure a code list in ATLAS.ti in our library: http://downloads.atlasti.com/library/Friese_2009-09_1.pdf http://downloads.atlasti.com/library/Woolf_2007-03_13.pdf Video Tutorials: http://www.atlasti.com/video/article/coding-basics.html http://www.atlasti.com/video/article/code-management-ii.html http://www.atlasti.com/video/article/the-margin-area-part-i.html http://www.atlasti.com/video/article/the-margin-area-part-ii.html
Code Reference
The selected code word is written next to the quotation bar in the margin. You can recognize codes by the yellow diamond symbol. The entry in the drop-down list in Figure 2: Codes in ATLAS.ti reads: test 1 {1-0}. The first number shows the frequency (how often the code has been applied). It gives you some information on the groundedness of a code, i.e. how relevant this code is in the data. The second number the density (how many other codes this code is linked to). Hence the code Test 1 has been used only once so far and it is not yet linked to any other code. To learn more about linking codes, read the entry on the Network View function and take a look at the Quick Tour.
MAIN CONCEPTS
18
Word Cruncher
The Word Cruncher counts all words in textual PDs. The count can be limited to one PD only. To clean up the count, a stop list can be defined to exclude special symbols or words like and, or, the, etc. The result can be displayed in an Excel table. See the chapter on Special Tools in the full manual.
Memos
Memos capture your thoughts regarding the text and are an important device for creating theory. A "memo" is similar to a code, but usually contains longer passages of text. A memo may "stand alone" or it may refer to quotations, codes, and other memos. They can be grouped according to types (method, theoretical, descriptive, etc.), which is helpful in organizing and sorting them. Memos may also be included as the objects of analysis by assigning them as PDs. See the section on memo writing under Textual Level Basic Functions, p. 128 ff. in the full manual. Also read these library articles: http://downloads.atlasti.com/library/Friese_2009-09_1.pdf http://downloads.atlasti.com/library/Friese_2008-12_8.pdf
MAIN CONCEPTS
19
Families
Families are a way to form clusters of PDs, codes, and memos for easier handling of groups of codes, memos, and PDs. Primary Document families can be regarded as attributes or variables. Families can be combined using logical operators similar to codes and Super Codes. For more detail, see the chapter Family Life in the full manual. Network Views
Network Views are a bit more sophisticated than families. They allow you to conceptualize the structure by connecting sets of similar elements together in a visual diagram. With the aid of Network Views you can express relationships between codes, quotations, and memos. PDs, families and even Network Views can also be nodes in a network view.
MAIN CONCEPTS
20
Network Editor
The Network Editor displays and offers all editing capability to construct and refine semantic networks. In addition, it allows the visual creation and traversal of hypertext structures.
Relation Editor
When the built-in relations that are used to connect objects in Network Views are not sufficient, you can edit them or create new ones using the Relation Editor.
Link Managers
The Link Managers provide an overview of all code-code links and of all quotation-quotation links you have created. The full manual has further information on the network function under The Conceptual Level Networks, starting on page 211. Video tutorials: http://www.atlasti.com/video/article/network-building-i.html http://www.atlasti.com/video/article/network-building-ii.html http://www.atlasti.com/video/article/network-building-iii.html http://www.atlasti.com/video/article/hyperlinks.html
Analysis
Query Tool
For more complex search requests, the Query Tool is at your disposal. Here you can formulate search requests that are based on combinations of codes using one or a combination of 14 different operators (Boolean, semantic and proximity operators). See the chapter on Special Tools in the full manual. Video Tutorial: http://www.atlasti.com/video/article/the-query-tool.html
Super Codes
A Super Code differs from a standard code. A standard code is directly linked with the quotations to which it is associated, while a Super Code is a stored
MAIN CONCEPTS
21
query, thus provides an answer to a question (in the best case) that typically consists of several combined codes. See the chapter on the Query Tool in the full manual.
Codes-Primary Document-Table
The Codes-Primary-Document Table contains the frequency of codes across documents. It can be exported as Excel compatible table. Optional, the table cells can also contain the word counts for the quotations per code across documents. This option can be found under the Codes / Output menu.
MAIN CONCEPTS
22
Team Tools
User Administration
Manage the ATLAS.ti user database through the user administration tool. This is a prerequisite for collaborative work, but is also useful to individual users through personalizing the login or protecting an HU with a password User Management options can be found under the Extras menu. See also the full manual, the chapter on User Management).
Coding Analyzer
The Coding Analyzer is helpful after merging HUs. It tests for redundant codes, i.e., codes used more than once for data segments that overlap or are embedded in one another. You find it under the Tools menu.
MAIN CONCEPTS
23
Export
RTF
There are output options for each of the main objects in ATLAS.ti: Primary Documents, Quotations, Codes and Memos. You find it under the respective menus. You can output a list of primary documents along with their comments, all or selected quotations with our without comments, quotations by codes (via the Codes / Output menu), a list of all codes, code-neighbors lists, etc. For memos, you can either output just a list of your memos, the content of memos or the content including the linked objects.
PDF
You can print out coded documents as you see them on the screen with the codes in the margin area. This option will soon also be available for PDF and image files (in a version higher than 6.2.23). You find it under DOCUMENTS / OUTPUT / PRINT WITH MARGIN.
XML Generator
The XML generator exports all information contained in an HU in XML format. This universal, open data format allows for a wide variety of possibilities for display, processing, and even integration of your data with external applications. Click on the XML button in the main tool bar to explore the options. Further detail is provided in the full manual and on our website: http://www.atlasti.com/xml.html XML Converter The XML Converter lists HUs saved in XML format and applies so-called "style sheets" (miniature programs written in XSLT) to the XML files. The user may edit the supplied style sheets or add new ones. Just to mention a few outputs that you can create using the XML style sheets:
A nicely formatted code book (Modern codes (commented only) A tag cloud for the elements in the HU (HU tag cloud) A list of quotations sorted by documents (Quotations List by Primary Documents) The content of quotations sorted by codes (Print Codes in Full)
Use version 1 style sheets in ATLAS.ti v.5.x and version 2 style sheets in ATLAS.ti v.6.x.
MAIN CONCEPTS
24
SPSS Export
You can export your coded data as SPSS syntax file. When executed in SPSS, your quotations become cases and your codes and code families variables. In addition further identifying information in form of variables is provided like the primary document number for each case, start and end position and creation date. These variables allow you to aggregate your data in SPSS if needed. If you need a less detailed output, see Code-Primary-Document-Table. The table provides an output that is already aggregated by primary documents.
Excel Export
The frequency of codes across documents (Codes / Output / Codes-Primary Document Table), the results of the Cooccurence Table Explorer (Tools / Cooccurence Tools, and the word frequency count (see Word Cruncher) can be exported as Excel compatible file.
Graphic Files
Network Views can be saved as bmp or emf files: Open a network view and select Network / Save as Graphic File.
HTML Generator
Publishing online or just creating a printout from a project is available with the HTML generator. The results can be viewed platform independent with any Web browser. This option can be found under the Extras menu and is best suitable for exporting your entire project. HTML preferences can be set under Extras / Preferences / HTML Preferences.
Miscellaneous Tools
Object Crawler
With the Object Crawler, you can search all of the parts of your project within ATLAS.ti that contain text. Searches are not restricted to just textual PDs: codes, memos, quotations, all families, code-code links, hyper-links, and the HU can be searched. In addition, the scope of the search can be limited to certain fields. You find the Object Crawler under the Tools menu.
MAIN CONCEPTS
25
Object Explorer
The Object Explorer displays all the elements of an HU in a strictly hierarchical manner, even if the structures are non-hierarchical, or even cyclic. You find the Object Explorer under the Tools menu.
26
A first important but often neglected phase is data and project management. The first step is data preparation. You find more information on supported file formats in the section Preparing Documents in the full manual. Apart from analyzing your data, you also manage digital content and it is important to know how the software does it (See the section How ATLAS.ti handles documents on page 93ff in the full manual. We also advise to read the chapter on project management, p. 270ff). In order to give you an idea what kinds of issues might come up, here is a list of issues that are important to consider: Where will you store your data? Do you want to work with external or internal documents? Do you want to link your data transcripts to the original audio and video sources? (Version 6 manual addendum) Do you think you might have to move your project at some point? (See Project Maintenance, p. 282ff).
27
Should data be edited after it is assigned to ATLAS.ti? See the section on Editing primary documents in the full manual, p. 78ff.) Also: http://downloads.atlasti.com/library/Friese_2009-09_2.pdf Do you need data to be shared among team members? For further information on team project management see the chapter on Collaboration (p. 262ff) and Merging Hermeneutic Units (p. 289ff). Where and how can you create project backups? See Project Maintenance, p. 282ff.) The One-Folder Setup - For Single Users and Teams Storing all data files to be analyzed and the ATLAS.ti project file in one folder is the easiest and most flexible way to set up a project. Even when working with video files or other sizable data sets that you may store on an external hard disk or server, you can still use this setup. You just need to save the ATLAS.ti project file (the HU) on the external disk or server as well. This scenario also works well for team projects. Each team member either stores the project folder with all of the data on his or her computer, or the project folder is kept on a server. In the team project case a few more issues need to be considered, like working with user accounts and different sub-HUs. Teams can follow the instructions for the sample data set and then continue to read the additional information for team projects.
Working with Internal Documents (for smaller projects containing only text files)
Using version 6.2 or higher, it is possible to create internal documents. This means you no longer have to worry about document references and document locations. Your ATLAS.ti project consists of a single data file, the HU. This project setup can be used for smaller projects where the data material is comprised of text documents only. As this function was recently added to the software, it is difficult to know yet how small is small. A test project containing 40 interviews that amounted to a total size of 5.3 MB and 1800 pages of text outside ATLAS.ti, resulted in a coded HU file of 730k. Thus, working with 100 internal files should still be ok. This approach is also a good option for working with sensitive data material. When saving the HU, the data material is highly compressed and therefore unreadable outside ATLAS.ti. In addition, the HU can be password-protected. This should make unauthorized access reasonably difficult. This is how it works: you generate a new empty text document within ATLAS.ti and then you copy and paste the contents that you want to analyze into it: Open ATLAS.ti and select the main menu option DOCUMENTS / ASSIGN / NEW TEXT DOCUMENT.
28
Enter a name for the document into the window that opens. The document is loaded in edit mode. Copy the text you want to analyze and paste it into the primary document. Protect the document by quitting edit mode. You will find the edit mode button in the second toolbar below the main menu on the left hand side. Click on the pen and select the option Save and Leave Edit Mode. Proceed in the same way with all other documents that you want to analyze. Save the HU file. Make sure that you always create a backup copy of this file and store it at a safe location! See Assigning and Handling Primary Documents in the full manual for further details, and http://downloads.atlasti.com/library/Friese_200909_2.pdf from our library. - Video tutorial: http://www.atlasti.com/video/article/assigning-documents.html
The Process
There are two principal modes of working with ATLAS.ti, the data level and the conceptual level. The Data Level includes activities like segmentation of data files; coding text, image, audio, and video passages; and writing comments and memos. The Conceptual Level focuses on querying data and model-building activities such as linking codes to networks, in addition to writing some more comments and memos.
Data-Level Work
Data-level research activities include segmenting the data that you have assigned to a project into quotations, adding comments to respective passages (note-making/annotating), and coding selected text passages or data segments, secondary materials, annotations, and memos to facilitate their later retrieval. The act of comparing noteworthy segments leads to a creative conceptualization phase that involves higher-level interpretive work and theory-building. ATLAS.ti assists you in all of these tasks and provides a comprehensive overview of your work as well as rapid search, retrieval, and browsing functions. Within ATLAS.ti, initial ideas often find expression through their assignment to a code or memo, to which similar ideas or text selections also become assigned. ATLAS.ti provides the researcher with a highly effective means for quickly retrieving all data selections and notes relevant to one idea.
29
Read more detail in the section on the Data Level Basic Functions in the full manual.
30
Query the data based on your research questions utilizing the different tools ATLAS.ti provides. The key words to look for are: simple retrieval, complex code retrievals using the query tool, simple or complex retrievals in combination with variables via the scope button, the cooccurence explorer (tree and table explorer), the codes-primary document table, Excel and SPSS export of frequency counts for further statistical analysis. Build semantic, prepositional or terminological networks from the codes you have created. These networks, together with your codes and memos, form the framework for emerging theory. Look for the chapter Network Views in the full manual.
Finally, compile a written report based on the memos you have written throughout the various phases of your project and the networks you have created and even publish your project as a World Wide Web document or an individual presentation using XML. See HTML Export and Creating Reports with the XML Converter in the full manual. For your convenience a number of style sheets are already provided. Click on the XML button in the tool bar to explore them. See the following online resources for examples on how to work with ATLAS.ti: http://downloads.atlasti.com/library/Casasempere_2007-07_11.pdf http://www.qualitativeresearch.net/index.php/fqs/article/viewArticle/1632/3149 http://atlasmalaysia.wordpress.com/2010/09/01/howtolr/
PROJECT MANAGEMENT
31
Project Management
Project Setup
Create a new folder somewhere on your computer / server / external drive. Let's call it your ATLAS.ti project folder. Move a copy of all data you want to analyze into this new folder. If you have lots of data, you can create sub folders within the main folder where you store the data (see Figure 5: One-folder setup and Figure 6: One-folder setup with sub folders below). Open ATLAS.ti. Check the title bar. If the last used HU is open and not a new one, select the main menu option FILE / CLOSE. The text in the title bar should read: New Hermeneutic Unit. Select the main menu option DOCUMENTS / ASSIGN / ASSIGN. A file loader window opens. Navigate to your project folder. Select all documents in the folder. To select all documents, you can use the key combination Ctrl+A. Click on the open button (it will appear in the language of your Windows system).
PROJECT MANAGEMENT
32
Open the the drop down field for P-Docs and left click each document to find out whether you can load the documents on your computer. Documents are assigned in alphabetical order as listed in the file manager. In case the alphabetical order is not useful for your purposes, or if you do not assign all the data at once, you can always drag and drop a document to a different location. It may not always be possible to know from the very beginning what might be a good analytical name - or perhaps you have already created a project before reading these suggestions. In that case, you have the option to rename each primary document (right click on a primary document in the P-Docs Manager and select the option Rename). Renaming a primary document does not affect the data file names on the hard disk. It is only an internal change reflected in the HU. Thus, the entries in the Name column in the P-Docs Managers will differ from the entries in the column Origin when you change the name of a primary document.
PROJECT MANAGEMENT
33
Leave the P-Docs Manager at this position and proceed to save. Select the main option File / Save (or Save As) A window opens where you can specify the location of the HU file and its name. The default location is the TextBank folder. DO NOT save the Hermeneutic Unit there!!! Navigate to your project folder. The folder appears empty, as only files of the type *.hpr6 are shown. You can see this at the bottom of the window, where there is a field for file types. Enter a name for the Hermeneutic Unit, your ATLAS.ti project file. Click on the Save button shown in the language of your Windows system. Now, take a look at the column Origin in the P-Docs Manager again. If you followed the instructions correctly, then all absolute path references are replaced with the entry <HUPATH>.
Figure 7: HUPATH entries in the column "Origin" in the Primary Document Manager
Congratulations. That was probably the biggest hurdle in terms of project management in ATLAS.ti. The project can now be moved to any locationyou simply have to move the entire folder.
Working in Teams
When working in a team, you can set up your project in the same way as recommended above. What may be different is the location of the project folder, depending on the nature of the project. Basically, there are two possibilities: to store it on a server that each team member can access or, if access to a server is not always possible, team members can also store the project folder on their personal hard disks or external storage device. The one folder setup is the optimal solution for all these scenarios. In the following section, only the technical aspects of data management are discussed, such as the role and tasks of the project administrator and those of the team members. In addition, members of a team project need to discuss a number of further issues like how to develop the coding system, how to divide the work, how and when to inform the others about what one has done, how to share the analysis, etc.
PROJECT MANAGEMENT
34
Project administration
When working in a team, it is best to nominate one person to be the project administrator. Unless everyones skill level is equal, choose the person with the greatest knowledge of ATLAS.ti and the highest degree of computer literacy. The job of the administrator is to setup the project, to distribute it to team members, to provide instructions for team members, and to collect the sub projects from time to time in order to merge them.
Logging in
From the main menu select EXTRAS / LOGIN. Select your user account. Notice that you are not required to enter a password and that the default setting is Automatic Login.
PROJECT MANAGEMENT
35
After logging in, you are greeted by the software. ATLAS.ti will remember the setting and there is no need to log in every time. The automatic login can remain activated as long as the same user always works on the same computer. Only when different people use ATLAS.ti on the same computer at different times does this option need to be deactivated; in this case, the login window comes up each time you launch the software.
PROJECT MANAGEMENT
36
PROJECT MANAGEMENT
37
safeguard, double check the project name in the title bar after opening it, or always open ATLAS.ti first. The default setting is for ATLAS.ti to open the most recently used project, so you can be sure of working on your actual project instead of a backup of version.
When you reduce the auto recovery interval to less than 20 minutes, keep in mind that saving the HU in the background uses up computer resources. If you experience problems with the auto recovery backup, check the backup path and make sure you have sufficient space and rights to store files at this location.
PROJECT MANAGEMENT
38
A copy bundle file can also be used to freeze a certain state of your analysis, e.g. your first stage of coding. When writing the methodology section for your thesis or the research report, you can review the frozen stages to remind you how your analysis progressed over time. Copy bundle files are also useful for transferring projects between computers or for sending them to others via email. Then you need to unbundle or (as ATLAS.ti calls it) install the file (see below). When you use the HUPATH setup, you can also use WinZip or WinRAR to create a compressed file of your project. Use whatever you are comfortable with.
In the upper right section of the window you will see the list of all primary documents that can be bundled.
PROJECT MANAGEMENT
39
To remove documents from the bundle, double click on a document to move it to the left hand side. Or select all documents to be removed and click on the button with the errors pointing to the left. Next, you need to choose a location for your Copy Bundle file. The default file name is the name of the HU. The extension is .acb, for ATLAS.ti Copy Bundle. You can give the copy bundle file a different name, e.g. adding the date or a version number to the project name. After a short time a message pops up letting you know that the Copy Bundle file is ready (Copy Bundle finished and saved). Do NOT save the .acb file in your project folder. Save it to an external drive, a network drive, in the clouds anywhere it will be safe. If you store both your project folder and the .acb file on the same hard disk and it fails, you will lose all your work.
PROJECT MANAGEMENT
40
folder structure so that the unbundled HU can load the documents. Working with absolute path references is not a desirable project setup and should be avoided if possible.
Choose a location for the HU file in the field HU PATH. Check the color of the boxes in the column Use to see if there are conflicts and correct them if needed. Green and yellow are OK: if you are transferring a project for the first time or into an empty folder, you will see green boxes. When migrating a project between computers multiple times, then the boxes are yellow. This means the files already exist on the target computer and will be overwritten with newer versions. If no conflicts are reported and nothing needs correcting, click on the button UNBUNDLE. The acb file is then unpacked, and all the documents will be copied into the corresponding directories. The unpacked HU opens in a new HU editor window.
Possible conflicts
If you see a pink box, ATLAS.ti has noticed that there are multiple HUs accessing the same file, but they have different information for this document in terms of file size or date of last saving. Unbundling this document would
PROJECT MANAGEMENT
41
mean that the other HUs could no longer access it. It is possible to install such documents in RESTORE mode. If a necessary path does not exist on the target computer, it is usually created by ATLAS.ti. But there are situations when this cannot be done. Then you see a red box. This happens, for instance, when the reference of a document points to a network drive like an H: or L: or Z: drive that does not exist on the target computer; or if a user does not have access to a specific drive. In these cases, the path can be mapped. Click on the button MAP PATH. Then another window opens. On the left hand side enter the path that produces a conflict; on the right hand side enter a path that exists on the target computer to use as location for installing the file.
Figure 13: Mapping a path if the location does not exist on the target computer
A light green color indicates that the HU can access the document, but not via its original path reference. This happens when an absolute path reference was used in the first place. At some point all documents were moved into the same folder as the HU file, but the original path references were not changed to reflect the new setup (see below, Modifying the project setup). If you do not optimize paths after changing the project setup to the HUPATH, then ATLAS.ti first follows the original path reference. When it does not find the files there, it checks the folder where the HU file is located as an alternative or fallback location for the files. When light green boxes are displayed, you can still unbundle the file. But it is advisable to optimize the path references immediately after unbundling: To optimize the path references, select Documents / Data SOURCE MANAGEMENT /OPTIMIZE PATHS.
PROJECT MANAGEMENT
42
PROJECT MANAGEMENT
43
Load the document. Repeat this for all other primary documents that you want to change. This process is not always foolproof; sometimes the first half page will be filled with nonsense characters. But this can usually be solved by going into edit mode and deleting those lines.
What is changed?
If you modify a document that is linked to the HU, then you modify the original document on the hard disk. Thus, the changes are not only visible inside ATLAS.ti. If you open the document in Word or elsewhere outside ATLAS.ti, the changes are also visible.
PROJECT MANAGEMENT
44
be made to the coded segments. If the HU were to load such a document, the codes would no longer show up where they were supposed to. In order to avoid such a situation, the HU controls the size and modification date of a document before it is loaded. If a mismatch is found, the document is not loaded and you see only a blank HU editor. As it would be very restrictive if ATLAS.ti did not allow you to rescue such a document, there is an option to reset the last access information and this forces ATLAS.ti to load the document (under Documents / Data Source Management). However, this does not mean that your codes will magically appear where they are supposed to. If you have modified a document outside ATLAS.ti, it is up to you to check all codings and adjust them manually if necessary. But this is better than starting from scratch and recoding the entire document. When setting up projects using the new data handling methods in versions later than 6.2, you can no longer accidentally edit an assigned file outside ATLAS.ti, in which case, rule No. 1 becomes obsolete.
PROJECT MANAGEMENT
45
<TransformMod oldSel="1@10,1050@10" newSel="1@10" afterOldSel="1051@10" beforeOldSel="1052@9"/> <TransformMod oldSel="1@10,1@10" newSel="1@10" afterOldSel="2@10" beforeOldSel="1052@9"/> <TransformMod oldSel="1@10,1@10" newSel="1@10" afterOldSel="1@11" beforeOldSel="1052@9"/> </LOGS> As a user, there is no need to understand the content of a log file or even to open it. You just need to know that it is there and that its best not to touch it. If you delete a log file and it cannot be rescued from the recycle bin, you have lost the coding for this document. The only option that remains then is to disconnect the document from the HU, newly assign and recode it. By the way, deleting a log file is the one single activity that will cause permanent data loss (apart from losing your HU file of course). All other problems related to data management can generally be solved. As for rule No. 1, if setting up projects using the methods of data handling that will be implemented soon, you no longer have to worry about log files as they are no longer visible and users cannot touch them. Rule No. 2 then also becomes obsolete.
How to edit
By default, all documents are loaded in write-protected mode. To enter edit mode, Load a primary document that can be edited. When loading such a document, a second toolbar appears underneath the main toolbar. Either select the main menu item EDIT/DOCUMENT ACCESS/ENTER EDIT MODE or click on the edit button in the second toolbar. All edit options in the second toolbar can now be used. Try out the various options like enlarging or reducing the font size, changing colors, putting text into bold or italic letters, using the highlighter, etc.
PROJECT MANAGEMENT
46
data and the HUs on a server, then ATLAS.ti controls the process. Thus, you do not need to do anything more than instruct each team member to leave the log files where they are, and that they should not modify, delete or move them. If your team members are working at different locations, then editing needs to be tightly controlled.
PROJECT MANAGEMENT
47
Coder D is allowed to edit PD 31 to PD 40 When it is time to merge the various sub-HUs, Coder A sends the project admin his HU file project name_coder A.hpr6 and the rtf and log files for PD1 to PD10 Coder B sends the project administrator her HU file project name_coder B.hpr6 and the rtf and log files for PD11 to PD20, and Coder D sends the project administrator his HU file project name_coder D.hpr6 and the rtf and log files for PD 31 to PD40. The project adminstrator copies all four sub-HUs into one folder, along with all 40 rtf files and all 40 log files. Then she opens each sub-HU and synchronizes it. ATLAS.ti will recognize modification when merging HUs and goes through the synchronization process, but we recommend to update each HU and check it yourself before beginning the merge process.
48
In ATLAS.ti you can often use a number of different routes to achieve the same result. Once you see that you can use any one of three options - the main menu, the context menu or the buttons on the toolbar - you can use whichever suits you best.
Open Coding
You may have heard or read about open coding related to grounded theory. In ATLAS.ti open coding simply means creating a new code. And this is how you do it: Load a text documents and select a segment with the mouse and right click on the highlighted area. Select the option CODING/OPEN CODING from the context menu. OR: Select CODES/CODING/OPEN CODING from the main menu. OR: Select the OPEN CODING button from the vertical toolbar on the left hand side of the screen (see left).
Figure 15: Selecting the Open Coding option via the context menu
49
Take a look at your screen. With one click, you have created four new entries:
You have created your first quotation and code. Each is displayed both textually (in the respective drop-down list) and graphically (in the margin area). The separation of quotation and code has a number of advantages. As they are independent objects, you can comment on each code but also on each quotation. Quotations cannot only be linked to codes, but also to each other and to memos. This is a prerequisite for the hyperlink function discussed in the respective chapter in the full manual. It allows you to work directly at the data level without necessarily using codes.
As you can see in Figure 17: Meanings of coloured bars in the margin overlapping codes are displayed in different colors in the margin area. These colors help you to distinguish which bar belongs to which code word. If you were to add even more layers, the next two layers would be displayed in violet and blue. After that, silver is used again. The colors do not contain any information about the different code levels in the sense of higher or lower ranked categories. When you delete a quotation, the colors may change. It is possible to set your own colors for codes, but not for the bars marking the length of a quotation (see below Coloring codes).
50
There is no need to drop the code into the highlighted area. It just has to be dropped somewhere on the left hand side as shown in Figure 18: Drag & Drop coding.Whatever is highlighted gets coded.
Replacing a code
If you want to exchange a code for an already coded segment, then the easiest way is also to use drag & drop. This time, however, you cannot drop the code just anywhere; you have to drop it exactly on top of the one that you want to replace.
51
To enlarge a segment
Highlight an extended area by simply overwriting the existing one. Then click on the button REVISE QUOTATION SIZE in the vertical toolbar (see left).
Unlinking a code
Just as you can erase a pencil mark in the margin of a paper document, you can erase a code in the margin area on screen as well. The digital equivalent is called unlinking. It is mostly used when a segment is coded with more than one code. To unlink a code, right click on the code in the margin area and click on the UNLINK option. When you unlink a code from a segment that only has one code, the bar remains in the margin. If you want to erase both the code and the quotation, you need to proceed as explained below.
52
Quotation comments are visible in the bar in the margin area and by the tilde (~) sign in front of the quotation name in the quotations list.
53
Renaming codes
The renaming option is a global option. This means that renaming a code word in the Code Manager affects all coded segments that use this code. Mark a code word in the Code Manager, right click and select RENAME from the context menu. OR: Use the in-place way of renaming entries that you may know from working with the Windows file manager.
Coloring codes
If you want to color your codes, select a code in the Code Manager and click on the rainbow colored circle in the tool bar. To make code colors visible in the margin, right click on a white space in the margin area and select the option USE OBJECT COLORS. In order to still be able see which code label belongs to which quotation bar, a small bar colored in the same color as the quotation bar is shown in front of the code label. The code label shows the user-defined color.
Deleting a code
The delete option has global effects, it removes a code from the entire HU and from everywhere in your primary documents, code families or network views: Right click on a code and select DELETE. A message pops up:
These items are linked to code "[code name]: [x] quotations. Continue with deletion?
Confirm by clicking YES. An other message pops up asking you whether the linked quotations should also be deleted. If you dont need the quotations any longer, click the Yes button.
54
Merging codes
The Merge Codes function can be used to combine two or more codes with each other. This is a common procedure when you begin to clean up your code list after initial coding, or when you work in a team and merge sub projects. Begin with the code whose name you want to keep. Right-click and select the MERGE CODES option from the context menu. A small window pops up listing all other available codes. Select one or more codes from this list and click the OK button. To select more than one code, hold down the Ctrl key. A window pops up, telling you that the selected codes have been merged into the target code and that a comment leaving an audit trail of the merge process will only be entered if the codes already had a comment before merging.
Figure 20: Combined comments after merging leaving an audit trail shows the comment for the code Happiness after merging it with the two codes joyful and being content.
55
56
In addition to the Code Manager, open the Quotation Manager. Re-size the two windows, so they fit next to each other. Move the time line to a convenient position and expand it so it covers the full length of the screen. Set a filter to the currently selected document so that the Quotation Manager only shows the quotes of the video you are working on. From the main menu, select QUOTATIONS/FILTER/SELECTED PD. The drop-down list and manager show a pale yellow color, indicating that a filter is set.
Currently the time line cannot be scrolled. Thus, the entire video needs to fit onto it. Therefore the recommendation is to cut your audio or video files to a length of about 20 minutes. Otherwise, the visual presentation of the quotations becomes very tiny and difficult to set. In version 7, ATLAS.ti will offer a margin area as already available for most other data file formats. Then the above described coding procedure can also be use for coding audio- and video files. To create a quotation, move the cursor to a start position on the time line. Click the button Mark selection Start. Position the cursor at the end point of the quotation. Click on the Mark selection End button. Then click on the Create Quotation button.
57
A new quotation is listed in the Quotation Manager. The default name for audio and video quotations is the document name. Each quotation can, however, be renamed.
Unlinking codes
To unlink a code from a video quotation, right click on the white background. Then select the option UNLINK CODES.
58
If you prefer time rather then frame settings for quotation references, right click the bottom right hand side of the time line and select CHANGE TIME FORMAT. Then a small window pops up showing you the various options available.
FAMILY LIFE
59
Family Life
Just as codes describe sets of quotations, families cluster Primary Documents, Codes, and Memos. One important objective is to manage large amounts of objects by classifying them into subsets, e.g., all theoretical codes, all PDs from respondents of a certain age group or location, all memos related to a theme, etc.
FAMILY LIFE
60
Code families can be used to loosely group codes that belong together. Other than in Network Views where specific relationships between codes need to be defined, in code families it is not necessary to specify the ways in which codes relate to each other. Use memo families to sort, filter, and organize your memos.
Super Families
Super Families are combination of other families (including other Super Families). You can build complex families incrementally from existing families using a set of powerful Boolean operators. From a technical perspective Super Families function just like Super Codes (see Super Codes on page 84).
Common Procedures
Although the handling of families and their members is identical, their usage differs between types of families. In the following, the general aspects of and procedures for working with families are described. Thereafter, a description of specific characteristics of the different family types follows. All procedures for displaying, editing, and creating families are found in the menus of their object type (i.e., under the Documents, Codes, and Memos menus).
FAMILY LIFE
61
Family Managers and the procedures described in the following are the same for all family types.
FAMILY LIFE
62
Creating Families
Before using a family, it must first be created in the Family Manager. Open the Family Manager for PDs, codes, or memos. Click the button CREATE A NEW ITEM or select the menu option FAMILIES / NEW FAMILY. Enter a name when prompted and click OK. If you enter the name of an already existing family, you will hear an error sound. A message pops up letting you know that the entered name is not accepted. Now you can add members and write a preliminary comment describing the family. The number of items in a family is displayed behind the family's name in parentheses.
Adding Members
After a family is created, adding members is often the next logical step. However, the assignment and removal of items can be done at any time during the life cycle of a family. To add new items to a family: Select a family. If you have just created the family, it is already selected. Select one or more items in the list pane on the right hand side. Click the Add Items button to make the selected items members of the selected family. If you want to add one item at a time, you can simply double-click an item.
Watch the member list being populated with the items you have just assigned. Another way to add members is to drag them from an Object Manager, the Object Explorer, or the margin area into the members pane.
FAMILY LIFE
63
Removing Members
Open the Family Manager for the relevant object type. Select a family. Select one or more items in the list pane for members. Click the Remove Items button to remove the selected items from the selected family. The items are only deleted from this family and not from the system. If you want to remove one item at a time, you can simply double-click on each item. If you delete objects that belong to a family somewhere else, they will be removed automatically from all their hosting families.
FAMILY LIFE
64
Figure 27: Double clicking on a family in the family manager activates a filter
If a family is activated as a filter, it is displayed in bold letters in the Family Manager. In the figure above, the cody family Role of the media has been set as filter. In addition, the background color in the Object Manager and dropdown list changes. The active filter is also indicated in the filter field of the status bar in the Object Manager.
To take the filter out double-click on the family again, or select EXTRAS / RESET ALL FILTERS from the main menu. A third option is to double-click on the filter field in the Object Manager while holding down the Ctrl-key.
FAMILY LIFE
65
Filters do not affect Network Views. Even if you have set a filter, Network Views still display all their nodes.
Removing Families
Open the Family Manager for the relevant object type. Select the family to be removed. Click the Delete button or select menu option FAMILIES / DELETE FAMILY. Removing a family does not remove any of the contained items.
66
Along with sharing all the characteristics of code and memo families, PD families have some additional characteristics.
PD families as variables can be used in queries and SPSS jobs.
PD Families can be used to define the scope of a query when used as global attributes supplementing codes. For instance, if the PD families "female" and "age group 1 (21-30)" were assigned to a number of interviews, one can then formulate queries like: "Show me all quotations from interviews with females between the ages 21 to 30 coded with "coping" or "power." For a step-bystep instruction, see Restricting Code Queries to Sub Groups on page 87. In SPSS jobs, PD families are a way to aggregate some of the data. PD-Family tables are an efficient means to create families and to assign PDs to their respective families. You can also use them to assign PDs and to generate variables in one step.
Use PD-Family tables to assign PDs and variables at the same time.
PD-Family Table
PD-Family Tables can be exported or imported as tab-delimited (XLS) or as comma/semicolon separated value (CSV) files. All three formats can be read by Excel as well as OpenOffice Calc. Below you see an example of an exported file. The gray colored columns have been added for readability. Below each
67
This data exchange format can also be used to conveniently assign a large number of PDs from generic lists. Documents: The PD sequence number. Name: The name of the PD. Path: The resolved path of the PDs data source, i.e., the actual location of the data source at the time the table was exported. @Origin: Contains the original path reference that was valid at the time the document was assigned to the HU. May contain special paths (e.g., <HUPATH>). This attribute can only be handled intelligently when importing such a table with ATLAS.ti. Other applications may simply ignore it and use the path attribute instead. All following columns: Families.
Type of Variables
Within ATLAS.ti, all families, when interpreted as variables, are dichotomous because an item may or may not belong to a specific family, thus it I encoded with either 0 or 1 / applies or does not apply.
By following a simple naming convention, PD families can be turned into nominal and categorical variables for the use outside of ATLAS.ti in statistical and other database applications. Let's assume that your respondents come from four different locations. To represent this in ATLAS.ti, you need to create four PD families, one for each location. Exporting these for families as PD-Family table to Excel, you get a table as shown above containing one variable per family with the values 0 or 1. If you prefer just one variable for location rather than four, you need to use the following naming convention when creating your PD families: VARIABLE NAME::VARIABLE VALUE:: Location::Berlin Location::London
68
Location::New York Location::Tokyo When exporting the PD-Family table, the result will be as follows:
The table shows one variable for Location and the respective values as string: Berlin, London, New York and Tokyo. Note that the variable name is prefixed with a hash sign (#). This is the syntax that needs to be used when preparing a table for import (see below). The hash sign indicates to ATLAS.ti that this variable needs to be turned into multiple families.
69
Do not assign a PD to more than one family with the same variable name (e.g., both to Location::Berlin and Location::New York). ATLAS.ti allows this, as it treats every family as a separate dichotomous variable. However, when converted into a categorical variable Location in the process of creating a PD-Family table, such multiple assignments could create problems. How to prepare and import a PD-Family table, see the full manual (available from http://manual.atlasti.com)
70
Possible uses for memos: To take methodological notes As a "bulletin board" in team projects As a project planning device. As a primary document To create code lists
Writing memos is an important task in every phase of the qualitative analysis process. The ideas captured in memos are often the pieces of a puzzle that are later put together in the phase of report writing. Theory-building, often associated with building networks, also can involve the use of memos. Memos are explanatory and descriptive texts that may be associated with other "objects" like quotations, codes, or other memos. Memos can also "stand alone" simply as part of an HU. They can contain methodological notes; they can be used as a bulletin board to exchange information between team members; you can use them to write notes about the analytical process , keeping a journal of to-dos. Memos may also serve as a repository for symbols, text templates, and embedded objects that you may want to insert into PDs or other memos. Memos can also be assigned as primary documents, if you want to code them.
71
Comments are not displayed in browsers separately from the object to which they are attached. Memos can be associated with more than one object and have an additional type attribute, e.g., theoretical, methodological, commentary, etc. They can also be free-standing, unlike comments.
Memo Content
Memos may use plain-text as well as Unicode or Rich Text. The latter allows extended formatting and offers the option to insert pictures, tables etc. You can also include local file names like file://notation.
Below the toolbar, the editor includes a title and memo type field. The default title assigned to a newly created memo is: ME todays date. The title and the default type Commentary can be changed by overwriting the default title or by selecting a different memo type. Various options are available. You can change the default memo title and manage available memo types. You can also be prompted for a title when creating a new memo. After having typed or changed something in the memo editor and before closing the editor, save the contents if you want to keep it. Click on the Save button or select MEMO / SAVE from the editors menu.
72
Save the memo content before moving on to do something else by clicking on the Accept button, or select MEMO / SAVE or click Ctrl+S. If you forget to save it, you are reminded by the program to do so before closing the editor.
73
Overview
Analysis Tools
In the first part of this section, you find a description of exploratory and advanced analysis tools. These are: the word cruncher, the query tool, the cooccurrence tools, the code-primary-documents table, super codes and super families. Word Cruncher: Creates a word frequency count for textual primary documents (see page 75). Query Tool: Finds quotations based on a combination of codes like: Show me all quotations where both Code A and Code B have been applied. Such queries can also be combined with variables in form of PD families. Thus, you can restrict a query to a certain sub groups of your documents: Show me all quotations where both Code A and Code B have been applied, but only for female respondents between the age of 21 and 30 (see page 77). Co-occurrence tools: Different from the query tool, where the results are quotations, you can use the two co-ocurrence tools to find codes. The tree explorer list all codes on the first level of the tree and if there are more codes that have been applied to the same or overlapping quotations, then you can open the tree branches to see those codes. If you open the tree further, the associated quotations are shown. The second tool, the table explorer, shows a cross-tabulation of codes and within the cells a frequency count of how often each pair of codes co-occurs. The cells also provide access to the data behind them (see page 89). Codes-Primary-Documents-Table: This table is a cross-tabulation of codes by primary documents. It shows how often a code has been applied to a document (see page 95).
74
Super Codes are saved queries. They can be reused, e.g. after more coding has been done or after coding has been modified. They can also be used as part of other query, thus, you can build complex queries step by step (see page 84). Super Families: Like super codes, super families are saved queries based on families. The purpose is to create families on an aggregate level. For instance, if you have your base families for age groups, gender and location, you can create super families that reflect a combination of these base families like all females from age group 1 living in city X (see page 89). Intercoder-Reliability: To assess intercoder reliability for data coded in ATLAS.ti., you can use the web based tool CAT to calculate Cohen's kappa and Cronbach's (alpha) See the full manual for further detail (http://manual.atlasti.com).
75
You may specify a number of properties before starting the frequency count. Include Selected PD only: Select the scope of the count to be all textual documents or the selected one only. Use Built-in Tool: If you want to output the frequency count in Excel compatible format, uncheck this option. The built-in tool is a simple list
76
window. In addition to the frequency count, size and percentage for each word is shown. The built-in tool can only be used for a selected PD. Use Stoplist: Use a stop list to exclude words from the count. Instead of enumerating every variant of a word, you can specify patterns. You can create powerful stop lists for different languages, different topics, and different levels of detail. Clean text before counting ignore: The characters specified in this dialog will be removed from the source word before it is counted. Example: including " eliminates the difference between "Hello" and Hello. Ignore case: If checked, upper and lowercase letters are ignored: Hello, HELLO, and hello are not counted separately. After all options are set, click the OK button. The figure below shows the result when using the built-in tool for a single PD. If you want to see the word frequency count for all or selected documents, use the Excel output option.
Figure 35: Word Cruncher output when using the internal tool
77
Query Tool
You need the Query Tool for queries using more than a single code.
The Query Tool is used for retrieving quotations using the codes they were associated with during the process of coding. This is different from a text search: To search for occurrences of text that match a specified pattern or string, you have to use the search function or the Object Crawler (see the full manual, http://manual.atlasti.com). The simplest retrieval of this kind (search for quotations with codes) is what you frequently do with the Code Manager: double-clicking on a code retrieves all its quotations. This may already be regarded as a query, although it is a simple one. The Query Tool is more complex in that it can be used to create and process queries that include combinations of codes. A query is a search expression built from operands (codes and code families) and operators (e.g. NOT, AND, OR, etc.) that define the conditions that a quotation must meet to be retrieved (e.g., all quotations coded with both codes A and B). By selecting codes or code families and operators, a query can be built incrementally which is instantaneously evaluated and displayed as a list of quotations. This incremental building of complex search queries gives you an exploratory approach toward even the most complex queries.
78
The Query Tool has the following main components: [1] The operator toolbar, located near the left margin of the window. [2] The code-family pane in the upper left lists code-families to be used in queries. [3] The codes pane below the code-family pane contains all current codes (set filters do apply). [4] The term-stack pane in the upper right displays the stack of all expressions entered in the current query. If more than one entry is visible, there are arguments still waiting to be used in the query. The topmost entry is the current query. [5] The current query is also displayed in the feedback pane directly below the term-stack pane. Here a different notation is used, one that uses parentheses and resembles the calculator style of entering queries. [6] The result of the query is displayed in the results list located in the lower right of the window. Above the term-stack pane are several buttons for manipulating the stack: swapping (S) or duplicating terms (P), clearing the stack (C), etc. Close to the results list are two buttons for removing unwanted hits and creating a report.
79
In figure Figure 36: The Query Tool window you see three other buttons highlighted in green. A super code is a saved query (see Super Codes on page 84 for further detail). You need the adjacency operator settings if you want to search for codes near to each other . Behind the Scope button you find another important feature. When you click on the Scope button, a second window opens showing the PD families you have created. These are often variables like age, gender, education, profession, location, time intervals etc. The scope function allows you to combine a code query with variables. For instance you can ask for all quotations where you have applied code A and code B, but only for females between the ages of 21 to 30.
Operands
Basic Operands
Two sorts of basic or atomic operands may be used in a query: Codes and code families. A code represents a set of quotations, while a code family yields the quotations of all the codes that its members have. In other words, a family is interpreted as its member codes connected by the Boolean operator OR. Selecting a code family F1 which contains five codes C1-C5 is equivalent to the query: "C1 OR C2 OR C3 OR C4 OR C5".
Complex Operands
"Operand" does not only apply to basic descriptors. An operand can be any expression that itself is used as an argument. An expression "A AND B" may be used in a more complex query as an operand: "NOT(A AND B)", "(A AND B) OR (C AND NOT D)", etc. All types of operands can be freely mixed in a query using any of the operators described below.
Operators
Three sets of operators are available. They are located within the toolbar at the left edge of the Query Tool. Boolean operators allow combinations of keywords according to set operations. They are the most common operators used in information retrieval systems. Semantic operators exploit the network structures that were built from the codes. Proximity operators are used to analyze the spatial relations (e.g., distance, embeddedness, overlapping, co-occurrence) between coded data segments. You can display a short help message for each operator by right clicking on its corresponding button in the toolbar.
80
Build complex queries incrementally with immediate feedback after each step.
A Boolean Query
The example below uses the HU Jack the Ripper_stage II. Please load and display this HU while reading the following. You can access the samples file , via HELP / QUICK TOUR. Our sample query, using Boolean operators, is this: Find all quotations coded with either code reason released: alibi or code reason released: lack of evidence. Open the Query Tool by clicking on the binoculars button in the main toolbar. Double-click on the code reason released: alibi. The Query Tool displays the following entries:
81
The term stack and feedback pane now display the code reason released: alibi. The results pane lists all quotations for this code. Double-click on the code reason released: lack of evidence.
82
With two operands on the term stack, we can combine them with an appropriate operator. The intention was to retrieve all quotations that contain information about an alibi or lack of evidence as reasons to release a suspect. Click on the OR operator (see left) to combine the two expressions from the stack.
The term stack now contains only one term, OR(reason released:alibi,reason released: lack of evidence), i.e. the combination of the two codes. This term can be used as an operand to further extend the query, e.g. to negate the expression or add some more codes to it. But we will stop here for now. The feedback pane displays the query in infix notation, as we would have entered it into a regular calculator (reason released: alibi OR reason released: lack of evidence). The results pane lists 10 quotations. You can look at the quotations in the context of the document by clicking on a quotation in the list, or you can create a report.
83
Creating a Report
To print all hits found by a query, click the PRINTER button to the right of the results list.
Select one of the following report options: LIST: Print a list of all quotations in a compact format showing only the quotation names. LIST INCLUDE COMMENTS: Same as LIST but includes the quotations comments if any. FULL CONTENT: Output the complete text of the quotations. FULL CONTENT NO META: Output the complete text of the quotations and specify the kind of information that is included in the output. You may for example exclude the meta information for each quotation. FULL CONTENT INCLUDE COMMENTS: Same as FULL CONTENT but includes the quotations comments if any.
84
Then, choose whether the output should be displayed in a text editor, saved to disk, or printed (see also Output Destinations on page 125).
Super Codes
"Intelligent" Super Codes compute their quotations on demand. Super Codes are a convenient way to store queries. Super Codes are very similar in look and feel to normal codes, with one important difference: instead of hardwired connections to quotations, Super Codes store a query to compute their virtual references whenever needed. They automatically change their behavior during the course of theory building. If you have a Super Code All about Magic with a query SUB Magic and later add another sub code White Magic,, all quotations to which the latter code refers are also retrieved by the (unchanged) query of All about Magic. Super Codes can be clicked on in the code list like any other code and they will display their quotations in an identical way. Normal codes are hard-wired to their quotations.
85
Super codes are displayed in the Code Manager just like regular codes and can be recognized either by a red text color or by their red symbol, if images are switched on in the Views menu. The list of quotations associated with the Super Code can be displayed with a double-click, just as for any other code. Frequencies (density) are only indicated if you activated it, e.g. with a doubleclick in the Code Manager. If you start a new session, an asterisk (*) replaces the frequency count. The reason for this is that a Super Code is dynamic and its density/frequency count changes as soon as you modify any of the codes contained in the query of the Super Code. For the same reason, Super Codes are not displayed in the margin area. There is the possibility to create a regular code from a Super Code.
Super Codes can contain Super Codes can contain....
Super Codes can be used in code families, Network Views, and, last but not least, as powerful operands in queries, allowing you to incrementally build complex queries.
86
Enter a name for the new Super Code or accept the default name created from the query expression. Click OK. The newly created Super Code immediately appears in the list of codes and can be used for new queries (and Super Codes) right away. Its icon and code name are red.
You can access and edit the query later, but the notation that is used in the edit query window is not so easy to understand. Therefore it is advisable that you enter the query into the comment field of the super code. You can copy and paste it from the feedback pane into the comment field.
87
Figure 26 above shows a simply query for the code lang_direct quote (use of direct quotes in newspapers articles). The question of interest is whether this writing style is used equally across the various newspapers. The two countries compared are Germany and the USA. The frequency for all documents is 25 (see the code in the code list on the left hand side). The scope is set to the PD family Country::USA. This means that the result pane is filtered only displaying quotations from US newspapers. This results in 16 quotations, which can be seen at the bottom left of the query tool window. Double-clicking on the PD Family Country:: Germany in turn yields the remaining 9 quotations (16 + 9 =25). Further comparisons could be by circulation, local versus national papers, or by political orientation comparing papers that are more politically right or left oriented.
88
For the results always look at the result pane of the query tool window. The bottom pane of the Scope of Query window shows all quotations from the selected document group. In mathematical terms, the result pane in the query tool window shows the intersection between the code query and all quotations from a particular document group. You can also combine a number of PD families or individual PDs using Boolean operators in the Scope of Query Window (see Combining group of documents to restrict searchers below). A scope is not stored as part of a Super Code's query specification. When you process the query of a Super Code later, the complete data base is queried by default.
89
Super Families
Just like Super Codes, Super Families recalculate their members on demand.
Super Families follow the same underlying logic as Super Codes (cf. Super Codes on page 84). They are constructed by combining families or already existing super families. Their members are determined dynamically whenever you activate a Super Family. Super families can be created based on primary document families, code families and memo families. (For further detail, see the full manual, http://manual.atlasti.com).
Co-occurrence Tools
The Co-occurrence Explorer works on a different logic than the Query Tool. In the Query Tool the user has to determine and select codes or code families and the appropriate operator. The results are a list of quotations. The Cooccurrence Explorer by default looks for all codes that co-occur in the margin area combining the operators WITHIN, ENCLOSES, OVERLAPS, OVERLAPPED BY and AND without the user having to select a a specific code, code family or operator. It is however possible to work with filters, so that the tool is not always creating output for all data and all codes. The Co-occurrence Tools offer two output choices. You can either view the results in form of a hierarchical tree or in form of a table. Below you see an example for both.
90
The same option is available for primary documents. If you expand the branch
Figure 49: Expanding to code and quotation level in the tree explorer
for Primary Docs, you can see which codes have been applied to this PD. Further, you can expand to the quotation level to look at the material coded there. Let's take a look at a potential question that the Co-occurrence Tree Explorer can answer:
91
With one click you can see which codes were used when the codename of suspect: Kominiski was applied: the description of the suspect, his potential motive and a list of reasons why he was suspected. If you expand the tree one more level you gain access to the full context with a click on the quotation link. In the section Explaining frequency count and number of quotations listed on page 94 it is explained how to interpret the listed quotations. If you want a count of the number of quotations that co-occur, you need to run the table explorer (see below).
92
Using the default settings, a c-coefficient is immediately displayed . You find more information on the c-coefficient in the full manual (http://manual.atlasti.com) The entry n/a indicates that the pair of codes does not co-occur anywhere in the data material (= not applicable).
User-defined content
If you have not set a code family as filter, all co-occurrences for each pair of codes will be displayed. This potentially results in a huge table. Therefore, in the columns to the left of the table, you can select the codes that should be listed in the columns and rows. This produces more purposeful tables. In the first column, select the codes to be displayed in the columns using the usual Windows selection techniques for multiple item selecting via the Ctrl or Shift key. In the second column, select the codes to be displayed in the rows. Activate the option Use selected codes. You find this option above the two selecting fields.
93
Figure 52: Select codes to be displayed in either the columns or rows of the table
The table below provides a good overview of the reasons why the five suspects Cohen, Feigenbaum, Kaminski, Kelly and Kosminski were suspected:
Another way to reduce the codes to be displayed in the table is by setting a code family as filter before running the Co-occurrence Table Explorer.
94
Inspecting Content
In order to see the quotations associated with the two co-occuring codes, the table cell can be expanded as shown in figure 32. The quotations in the dropdown list provide a direct link to the data in context.
The number of co-occurrences in the above example is two, however three quotations are listed. Why this is the case is explained next.
95
Figure 55: Explaining the relation between frequency of cocccurence and number of quotations
Thus, there are three quotations, but only two co-occurrences that are counted for the frequency count.
96
overview of how often these codes have been applied to the various groups of documents (see figure ref below)
Figure 56: A possible outcome of creating Codes-Primary-Document tables for various PD families
In order to create such a table, you need four Super-PD-Families (see Super Families on page 89) based on the four base families for gender and location: female female male / male / / location A, / location B, location A, location B,
and a code family for the codes fear, worries, belief in technology, change of focus, no change and indifference (assuming that you have used a few more codes than just these seven). Then you create four tables and copy them all into one spread sheet. In most cases the totals are of highest interest and not the count for each individual PD. But this could of course also be the case, depending on your type of data and the kind of questions you are interested in. Based on this first result, you can aggregate the data further and could come up with the following comparisons:
97
98
Create the table by selecting CODES / OUTPUT / CODES-PRIMARY-DOCUMENTS-TABLE and then select the option you need: Quotation Count or Word count; the internal or the Excel table.
VISUALIZATION
99
Visualization
Network Views
The word "network" is an ubiquitous and powerful metaphor found in many different fields of research and application. Flow charts in project planning, text graphs in hypertext systems, cognitive models of memory and knowledge representation (semantic networks) are all networks that serve to represent complex information by intuitively accessible graphic means. One of the most attractive properties of graphs is their intuitive graphical presentation, mostly in form of two-dimensional layouts of labeled nodes and links.
VISUALIZATION
100
In contrast with linear, sequential representations (e.g., text), presentations of knowledge in networks resemble more closely the way human memory and thought is structured. Cognitive "load" in handling complex relationships is reduced with the aid of spatial representation techniques. ATLAS.ti uses networks to help represent and explore conceptual structures. Networks add a heuristic "right brain" approach to qualitative analysis.
Networks are more than just graphics!
Along with using networks for "mind mapping" and the visual design of theoretical models, you can exploit the structural properties of code-networks to enhance the retrieval of quotations. Using networks for retrieval purposes is a well known technique in information retrieval. In that domain, a structured sets of keywords is usually referred to as a thesaurus. This usage of networks for semantic retrieval is described in detail in the full manual (see http://manual.atlasti.com).
VISUALIZATION
101
the link starts and the target node is where it ends: the destination to which the arrow points. The terms connection and link are synonymous.
Links are created either implicitly (e.g., when coding a quotation, the quotation is "linked" to a code), or explicitly by the user. The latter option is described in detail in this chapter. Strictly speaking, code-quotation associations (codings) also form a network:
But you cannot name these links, the code is simply associated with a quotation through the act of coding. In a network view you can visualize these links. In
VISUALIZATION
102
ATLAS.ti all unnamed links are referred to as weak links, all named links are referred to as strong links.
A Network View is typically only a subset of this global structure of nodes and links combined with an individual layout of nodes. It is like viewing the same thing, i.e., the network, from different angles and with different pieces visible.
Node Types
The user can manipulate and display almost all objects within the HU as nodes in a network view: quotations, codes, code families, memos, memo families, other network views, primary documents (PDs), and PD families.
Relations
ATLAS.ti allows you to establish named links to more clearly express the nature of the relationships between concepts. With named links, you may express a sentence like "a broken leg causes pain" by two nodes (the source node "broken leg" and the target node "pain") connected with a named link ("causes" or "is-cause-of").
A default set of relations is provided. You may define your own.
The name of a link is displayed in the Network Editor as a label attached to the link midway between the two connected nodes. Six pre-set relations - or link types - are available in ATLAS.ti. These standard relations can be substituted, modified, or supplemented by user-defined relations. The default relations are listed in the table below. C1 and C2 are source and target nodes, respectively.
Relation
Label 1
Label 2
Width
Color
Formal Attribute
Layout Direction
C1 is-associated-with C2 C1 is-part-of C2
== []
R G
1 1
Black Black
Symmetric Transitive
VISUALIZATION
103
N A 0 P
1 1 2 1 1
Some of these characteristics directly affect the display of links, while others affect processing (e.g., search routines, automatic layout). A link between concepts is displayed in a Network Editor by a line with the relations label. You can choose from three different labels (label 1, label 2 and the name used for the menu when selecting a relation). The formal attribute affects both the display and processing capabilities of a relation. For example: All asymmetric relations are symbolized in the Network Editor with an arrow pointing toward the target code. Symmetric relations are displayed with an arrow at both ends. A typical transitive relation is the is-cause-of relation: if C1 is-cause-of C2 and C2 is-cause-of C3, it follows that C1 is-cause-of C3. Transitive relations also enable the "semantic retrieval" (see the chapter on the query tool in the full manual). The following properties are user-definable: he two labels and the menu text, which can be used as alternative display options, the width and color of the line linking two nodes, whether a link is directed or non-directed, and the preferred layout direction. The preferred layout direction affects the layout of a network when ATLAS.ti automatically arranges the nodes.
It is important to understand the difference between a relation (or a link type) and the link itself: There is only one "is part of" relation, but potentially many links using it. In the Network View below, the relation "consequence" is used only once, while the relation strategy is used four times. Another way to think of links and relations is to view links as instances of relations. Links are well informed about the characteristics of relations, which define their styles. If a characteristic of a relation is changed (e.g., line width, color, symbol), these changes are propagated to all links using it.
VISUALIZATION
104
Figure 61: Four links using two relations, strategy and consequence
The "Grounded Theory" method of Glaser & Strauss uses relations like "isphenomenon, "is-context-of, "is-consequence-of, "is-condition-for, "isstrategy-for," etc., to relate concepts found during the data-oriented open coding phase. In the analysis of argumentation structures, other relations are more suitable: e.g., "is evidence of, "is contradictory to, "warrants," etc. A medical expert attempting to capture diagnostic knowledge would use, e.g., "is-symptom-of, and "is medication for. You can create user-defined relations in the relation editor that you can access via the main menu's Network menu. (For further detail, see the full manual, http://manual.atlasti.com).
The Network Editor offers an intuitive and powerful method to create and manipulate network structures. It favors a direct manipulation technique: You can literally "grab" codes, quotations, memos, or other objects using your cursor and move them around the screen as well as draw and cut links between them. The following describes various methods available for creating and editing Network Views.
VISUALIZATION
105
The Network Editor has two "helper" windows ("NWE-Toolbox" and "Alignment Tool") that offer convenient support for some tasks. The Relation Editor for maintaining the relation "database," and for creating and editing relations is described in a separate chapter in the full manual
VISUALIZATION
106
Enter a name for the new Network View. A Network Editor opens. Import nodes with any of the methods described in the section Nodes below.
VISUALIZATION
107
Drag selections of a primary text: This method creates nodes from textual quotations or new selections of text. Open the PD that contains the text that you want to import into a Network View. Make a selection within the PD or select an existing quotation. Drag the selected text into the Network Editor. Release the mouse button at the position you want the new node to be placed. A new node is created and displayed. If a quotation did not yet exist for the selection of text, it will now be created. Drag text from other applications: This method allows you to drag text from drag & drop-compliant applications like Word into a Network Editor. Text dragged from Word is converted into a new memo and displayed as a node. Open the application from which you want to drag selections of text in to a Network Editor.
Text dragged from other applications becomes a memo. Text dragged from the PD pane becomes a quotation.
Mark the piece of text within the (other) application. Move the mouse pointer over the selected text. The pointer changes its appearance Hold down the left mouse button and drag the selected text into the Network Editor. Release the mouse button at the position you want the new node to be placed with its upper left corner. A new node is created and displayed.
VISUALIZATION
108
Select the objects to be imported into the Network View. Click Import. The imported objects are placed along the upper left corner of the Network Editor. You can either distribute them manually by moving each node with the mouse, or you can place them automatically by (mis-)using the option LAYOUT / SEMANTIC LAYOUT.
Linking Nodes
The links between nodes in a network are real connections between the objects. Therefore, creating and removing links should not be regarded as solely "cosmetic" operations. Links make permanent changes to the HU. There are several ways to link nodes:
VISUALIZATION
109
The two nodes are now linked to each other. In case of a first-class link between two codes or between two quotations, the relation name is displayed either within the link line, above the line (rotated) or within a box.
You can set the link display options under DISPLAY / LINK DISPLAY.
Cutting Links
Several approaches to disconnecting previously linked nodes are available. The first method works for all types of links and is useful when many nodes linked to one other node are to be disconnected: Select one or more nodes whose connections to another node are to be removed. Choose LINKS / CUT LINKS from the Network Editor's menu or click on the Cut Links button in the toolbar. Move the mouse pointer with the "rubber bands" to the target node. Click the left mouse button Alternatively: Click on one or more link labels.
VISUALIZATION
110
Choose LINKS / CUT LINKS from the Network Editor's menu or click on the Cut Links button in the toolbar. Or: Move the mouse pointer over a link label. Right click and choose CUT LINK from the context menu. The latter two methods work on "first class" links only: code-code or quote-quote (hyper) links.
Modifying Links
The type of a link (e.g., its Relation) can be changed in the Network Editor. Right click on a link label and select CHANGE RELATION from the context menu. The relation menu pops up. Select a different relation. A very efficient way to manipulate first class links is offered by the Link Managers that you can access via the main menu's Network menu.
Removing Nodes will not delete the objects they represent! The option Delete Node, however, does!!
VISUALIZATION
111
This procedure imports neighbors of all types. If a code with many quotations is selected, this operation might fill the Network Editor with unwanted quotation nodes. To suppress the import of quotations, hold down the CTRL key when issuing this command.
If you have mistakenly imported the wrong or too many node neighbors, select NODES /UNDO IMPORT NEIGHBORS from the menu, or press the key combination Ctrl-Shift-Z.
Creating Output
Several output options are available for Network Views. The range is from printing the layout (via copying to the clipboard in a variety of formats both textual and graphical) and by saving a Network View to a file.
VISUALIZATION
112
Printing Networks
Before printing a Network View the first time, you may prefer to set some general options (include title, print border, etc.) ( SPECIALS / PREFERENCES / PRINTING). You can either print the entirety or part of a Network View. Open the Network View and arrange all nodes to be printed. If you want to print a selection of nodes only, select these nodes (for multiple selection, hold down the Ctrl key). Choose NETWORK / PRINT NETWORK VIEW from the Network Editor's menu. In the Printer dialog window that opens, check "Selection" to print selected nodes only. If the size of the network view determined by the node layout exceeds a single page, depending on current printer settings, scaling or multi-page print will be offered as a choice. Make your choice and the network view will be printed.
Select all quotation nodes that should no longer be referenced by the original node. In the above example, number magic remains only linked to quotation (1:29), Magic 7 remains linked to quotations 1:2, 1:3 and 1:4. All other quotations are linked to Magic 3.
VISUALIZATION
113
Select LINKS / CUT LINKS from the menu or use the unlink button in the toolbar. Red rubber bands appear, connecting the selected nodes with the mouse cursor. Move the mouse pointer over code node that should not longer be linked with these quotations. This unlinks all selected nodes from the original code.
VISUALIZATION
114
Hypertext
A network with text (or other media) as nodes is often referred to as a hypertext. The original sequential text is de-linearized, broken down into pieces that are then reconnected, making it possible to traverse from one piece of data to another piece of data regardless of their original positions. The procedures described so far have focused on the creation of code networks. Direct linking of data-segments (quotations) to other data-segments offers similar flexibility in choosing and defining relations. Almost all of the editing functions described for code networks can also be used when connecting two or more quotations. Code-code and quote-quote links are the only types of network connections that allow you to assign a name to the connection that appears on the line or arrow that runs between the objects.
VISUALIZATION
115
Cross-references between text passages are very common even in conventional media like books - just think of religious and juridical texts, literature, journals etc. Footnotes and endnotes are another common deviation from the pure linearity of sequential text. However, in conventional media, not much navigational support is provided for "traversing" between the pieces of data that reference each other.
Today's largest hypertext system: the World Wide Web
Computer-related hypertext applications include, for example, online help systems that display operational information in suitable small chunks (compared to lengthy printed information), but with a considerable amount of linkage to other pieces of information. A well-known hypermedia structure is the World Wide Web with its textual, graphical and other multimedia information distributed world-wide.
Benefits of Hypertext
What are the advantages of direct connections between text segments, compared to the traditional procedures of qualitative text analysis?
VISUALIZATION
116
retrieved with the help of the attached code words. But this may not be the only way you want to look at your data.
Hypertext makes text-text relations explicit
The concept of hypertext introduces explicit relations between passages. These links have to be built manually and result from an intellectual effort. The system cannot decide for you that segment x is in contradiction to segment y. But after the work of establishing the links, you can make semantically richer retrievals: "Show statements contrary to statement x." Hypertext allows you to create different paths through the data you are analyzing. For example, you may create a timeline different from the strict sequence of the original text.
VISUALIZATION
117
General Procedures
Star or Chain connections
A hyperlink chain connects quotations sequentially
When linking quotations, you have the option to create a chain, a star or a combination of both. Below, a chain and star connection are illustrated.
When creating this chain, the quotation 2:3 served as a source quotation and was linked to the target quotation 3:5. In order to continue the chain, the target quotation 3,5 became the source quotation and was linked to the new target quotation 8:7.. A hyperlink star connects many quotations from one source quotation:
VISUALIZATION
118
When creating a star, there is one source quotation and multiple target quotations. In the above example, the source quotation is 8:6, which is linked to five target quotations via a number of different relations.
All hyperlinked quotations can easily be recognized in the Quotation Manager. All source quotations are marked with an opening angle bracket <, all target quotations with a closing bracket >. If a quotation is both, source and target (as the case when creating chains), then both brackets are used as prefix <>.
When working with hyperlinks, it is advisable to set the margin display options as follows:
VISUALIZATION
119
To open this context menu, right click on a blank space in the margin area.
Creating Hyperlinks
ATLAS.ti offers a variety of options for creating and traversing hypertext links. Similar to the linking of codes, you may create hyperlinks in a the Network View editor (see Linking Nodes on page 108). In addition, hypertext links can be created "in context", or via Drag & Drag in the Quotation Manager and in the margin area.
VISUALIZATION
120
Figure 74: Select a relation to link the source with the target quotation
Select the option CHAIN to link a next data segment to the current target quotation. Select FINI to end the chain.
Figure 75: Select "Chain" to link another quotation to the current target quotation
To create a star:
Proceed as explained above To create a chain:, but select the option Star. Then the next quotation is linked to the current source quotation. For your information, the quotation IDs for the current source and target quotations are listed in the menu.
VISUALIZATION
121
Figure 76: Select quotations, drag and click target quotation in the Quotation Manager
Traversing Hyperlinks
Media-type quotations can be distinguished easily by their icons. These icons may be used in the margin area, the Object Manager, the Object Explorer, the Crawler, and in Network Views. The figure below shows the media types of hyperlinked quotations in the margin. From top to bottom these are: text, video, audio, image and Google Earth hyperlinks.
VISUALIZATION
122
If needed, open up the properties context menu in the margin area and select OBJECT TYPES / HYPERLINKS. Double-click a hyperlink displayed in the margin. The quotation to the left of the margin area is highlighted and a pop-up window displays the hyperlink's contents.
To display the hyperlink in context, click into the pop-up window. Clicking outside the pop-up window cancels the process. When you double-click on a multimedia hyperlink in the margin, you do not see the usual info text popping up. Instead, the quotation is played directly without opening the Media Control window.
VISUALIZATION
123
Move the mouse cursor into the highlighted area of the document and click the right mouse button. From the context menu choose SHOW LINKS. Steps one and two can be combined by pressing the Ctrl key while clicking the right mouse button.
The quotation "on the other end" of the hyperlink is displayed in a pop-up window. To directly move to this quotation in context, click into the pop-up display, or click outside the pop-up to cancel.
GENERATING OUTPUT
124
Generating Output
This chapter provides an overview of all output options that are available within ATLAS.ti. In addition, some general procedures common to most output functions are explained.
Overview
ATLAS.ti offers numerous options to create output and reports. Most output is textual, but numeric or graphical output is also available where applicable (Network Editor). Typical textual reports include sorted and filtered lists of objects, like code, memos, families and quotations. Comments can often be included when needed. For textual primary documents a near what-you-see-is-what-you-get (WYSIWYG) output is available, including the annotations populating the margin area. Output options are available from the menus of the tools, e.g., the HU Editors menus, the Object Managers and Object Explorers menus. Some tools offer an output button, either as the only option (e.g., the Query Tool) or in addition to the menus (Code Manager). The target of an output is not necessarily restricted to the printer. Many of the text-related report procedures allow output to be directed to a text editor, the printer, or a file. Other kinds of output are not necessarily intended to be printed directly but are to be processed by other applications, like the SPSS export option, several Excel compatible outputs, and HTML and XML export. You can find those procedures in chapter Export & Import on page 130.
GENERATING OUTPUT
125
General Procedure
Output Destinations
When creating textual output or when creating text-based exports, the following dialog is usually displayed in the course of the process: You can send output to:
Editor: The output will be displayed within a text editor. This option is useful if modifications are needed before actually sending it to the printer or if you are unsure about the size of the result. Printer : If you select this option, the output will be printed (the printer dialog window opens where you may select the desired print options). File: If you select this option, the output will be saved to a file (a save file dialog opens). File & Run: If you select the last option the file dialog window opens. After it has been saved, the file will be opened by the application registered for its extension. (e.g., a CSV or xls file will be opened in Excel).
Report Layout
Common Header
When creating textual reports, a common header precedes the list of objects (e.g., all codes or all quotations for a selected code). The header includes information about the kind of report (All current quotations), the HU (name and file path), the current user, the date and time of the creation of the report and the current filter if any.
GENERATING OUTPUT
126
The figure above shows the report generated after setting the the quotation filter to Hyper-linked. The quotation filter is displayed in the header.
GENERATING OUTPUT
127
and output each of them at the same time. For example, you can highlight three codes within the Code Manager and create a report that contains quotations coded to each of the three codes. Another option to restrict output is to work with filters. The output options for the four main object types are explained in detail in the full manual. .
GENERATING OUTPUT
128
Figure 84: Select the option 'Apply Style Sheet' to make use of existing style sheets
Select a style sheet from the list displayed. Next you are asked which parts of the HU you want to convert to XML. If you select a style sheet that only shows codes with comments or a list of objects, then you do not have to include the full content of the primary documents.
An HTML file is generated as your output file. A file dialog opens offering to store the file in the ATLAS.ti XML folder. For the file to be displayed properly including all colors and images as defined by the style sheet, store the file at this location. You can accept the suggested default name or enter a different one. Then click the SAVE button. The output is displayed in your web browser (Internet Explorer by default; results may vary in different browsers).
GENERATING OUTPUT
129
Figure 86: Use the style sheet 'Modern codes' to create a code book
You can copy and paste this table from Internet Explorer into a MS Word document or other applications and, if desired, add some further formatting.
When exporting quotations from PDF documents, you may have to correct the formatting in Word and delete unnecessary hard returns. The original layout is of course lost as the table only contains plain text. As mentioned above, you can simply copy and paste the table(s) from the Explorer to a word processing program and edit them there.
130
This chapter provides an overview of procedures to export/import data and structures between ATLAS.ti and other software. The export procedures introduced in this chapter are not always clearly distinguishable from the chapter Generating Output on page 124. While the output options are generally display and print oriented, export of the HU or parts thereof allows for analysis of data and structures created with ATLAS.ti within other software packages.
SPSS Export
ATLAS.ti is intended primarily for supporting qualitative reasoning processes. On the other hand, especially with large amounts of data, it is sometimes useful to analyze the data in a quantitative manner using statistical approaches. ATLAS.ti provides an export function to permit further processing of the syntax file by SPSS, the Statistical Package for the Social Sciences.
131
Importing Codes
When you import codes in XML format, additional information like the date of creation and modification, the author and code comments (in plain text only) are also imported. This is how you do it: Select CODES / MISCELLANEOUS/IMPORT CODES (XML) from the main menu. A file dialog window opens. Select the XML file that you previously created and click the OPEN button. Code colors are not preserved in the XML file. Thus, if you import codes and their comments via the XML option, you loose the coloring.
HTML Export
The HU to XML export and conversion utility allows an alternative and very powerful and flexible approach to creating web pages for users willing to learn how to program style sheets. Using HTML Export and specifying the layout using the HTML generator on the other hand is simply a matter of clicking options without any need to learn HTML. Potential reasons for creating HTML versions of Hermeneutic Units include:
132
Create a web publication. Even intermediary results can be presented to other researchers without much effort. These reports are readable without needing ATLAS.ti on every system that has a web browser available. Create a report. A convenient way to create a printout of (almost) all objects contained in a Hermeneutic Unit using a web browser. ATLAS.tis HTML code generation converts the Hermeneutic Unit currently being worked with into an HTML document. The user can control the HTML creation entirely via the HTML Preferences dialog box. Of course, the generated code can be edited and refined with a plain text editor or specialized HTML tools if you are knowledgeable in HTML.
Select EXTRAS / EXPORT TO / HTML from the main menu. Select FILE & RUN as the output destination to immediately open the generated output in a web browser. Select EDITOR if you want to modify the source before storing it. Below you see an example of how an ATLAS.ti project in HTML format may look like:
133
134
Figure 89: Jumping to the network view section by clicking on the link in the table of contents
APPENDIX
135
Appendix
Useful Resources
Quick Tour
http://quicktour.atlasti.com To get started, we recommend to work your way through the Quick Tour. Within 1.5-2 hours, you get an overview of the main functions and an idea on how to conduct an analysis with ATLAS.ti. When you are ready to start your own project, use this Mini Manualas your continuing guide.
APPENDIX
136
Video tutorials
http://tutorials.atlasti.com If you like to learn via video tutorials, we offer a range of short videos on the following topis: Features and Interface, Getting started on a project, Coding, Visualizing Working with Network Views and Hyperlinks.
Library
http://library.atlasti.com The library contains a list of white papers by various authors (mostly ATLAS.ti trainers and consultants) extracted from past newsletters. Submissions are always welcome.
Knowledge Base
http://kb.atlasti.com Search the knowledge base for immediate questions you may have related to licenses, installation, update procedures, trouble shooting or use-related questions. A quick search may already provide the answer without having to dig for an answer in the full manual.
Social Media
YouTube
http://youtube.atlasti.com You find a growing number of video tutorials on the ATLAS.ti YouTube channel. So far videos are available in English and Spanish, other to come.
Facebook
http://facebook.atlasti.com Stay updated with the latest news on product updates, special offers, new training materials, etc. by joining us on Facebook. We are also happy to hear from users via facebook. Stop by and let us know about your projects and experience with ATLAS.ti!
Twitter
http://twitter.atlasti.com Follow us on Twitter for the most essential bits of information and announcements (version and service pack releases, special offers, training opportunities, etc.).
APPENDIX
137
Newsroom
http://newsroom.atlasti.com The newsroom aggregates all pertinent company news conveniently on a single platform. Press releases, articles and comments from the various ATLAS.ti profiles on social networking sites are agglomerated in real time. In addition, the newsroom lets you subscribe to available content via RSS so you're always up-to-date on what's going on with ATLAS.ti.
The options provided via the More Resources sub menu may change over time reflecting the up-to-date status on documentations available. So it may be worthwhile to check this menu from time to time. All options present links to internet web sites. Thus, you need to be online to access these resources. Apart from the Coding Analyis Toolkit (CAT), all other options are self-explanatory. Therefore, you find a brief introduction to CAT in the full manual (http://manual.atlasti.com).
Getting Support
http://support.atlasti.com The easiest way to contact the Support Center is via the main menu option HELP / MORE RESOURCES / CONTACT SUPPORT. Or access the Support Center directly via the above URL.
APPENDIX
138
From the categories provided, select the appropriate option so your request can be processed as quick as possible.
Forum
http://forum.atlasti.com ATLAS.ti maintains various a large forum for peer-to-peer communication, i.e. for users to (virtually) meet, exchange experiences with the tool, share tips and tricks, and discuss technical and methodological issues that arise with the use of the program.
APPENDIX
139
Troubleshooting
Reporting Bugs
Most errors that occur during the operation of ATLAS.ti are displayed and written to a special log file named ERROR.LOG. This text file is located in the user system directory. The contents of this file can only be interpreted by the developers of the program and is of no use for others. If you file a bug report, you might be
APPENDIX
140
asked to send a system report including this file to a specified support e-mail address. The error report contains the following: The ERROR.LOG file. The System Report (which you can view via EXTRAS / SYSTEM REPORT / DISPLAY SYSTEM REPORT) The users configuration file USER.INI. The ATLAS.ti configuration file ATLAS.INI. The list of files in ATLAS.tis program folder. The list of files in the users system folder. The information in the package contains technical information needed for the technical support personnel to analyze and respond to reported problems. Except for the content of the ERROR.LOG, the system report and two INI files, no other file content is transferred. To send an error report, from the main menu, select EXTRAS / SYSTEM REPORT /CREATE AND MAIL REPORT. A message informs you that the report has been created and which files it includes. Click OK.
Figure 92: This message informs about about the files included in the bundle
ATLAS.ti launches your default email client and creates a new message addressed to an ATLAS.ti support address. The report file (Report-name-yyyymm-dd.AtlRep) should already be attached to the mail automatically. In cases where it isn't, please attach it manually (its exact location is indicated in the onscreen message). The body of the message displays further instructions. Add a description of your problem in the mail body. If possible, name any circumstances under which the problem occurred.
APPENDIX
141
Make sure you have an active Internet connection. Send the mail. Our support team will usually respond in less than 24 hours.
APPENDIX
142
You have the option to prohibit the live update for instance for settings where users to not have administrative rights to the computer. Or if internet access is not always available. Then it can become quite annoying if ATLAS.ti starts the update request, just to find out after half a minute that no Internet access is available. You can set the proxy settings or enter an authentication if needed. We strongly recommend that you leave automatic update checking enabled (see Help menu). By always keeping ATLAS.ti up-to-date you avoid problems and benefit from ongoing continuous development of new tools and features. Also, in case you need to request assistance with a technical problem, our support team may not be able to assist you as quickly if your installation is severely out-of-date Updating regularly is beneficial on all levels, so the adage never change a running system we sometimes hear is--at least as far as ATLAS.ti is concerned--a misguided superstition that will hamper you more than anything. If it wasn't for the many (free) service packs/updates we publish for each program generation, users would still not be able to perform a great many tasks that they hence have come to consider essential for their work, such as survey processing, XML stylesheets, smooth PDF, and many more. All of these were introduced in free service packs!
APPENDIX
143
Glossary
Authorship
Authorship is essentially ownership. Any item created while a user is logged in will be tagged with the name designated for that user. By using filters, it is possible to compare the work of different authors in the same project. Some division of labor strategies will prevent tracing all work done by each person. Authorship can be changed globally or on an object by object basis.
Backup
Use the copy bundle function to backup your entire project: Tools / Copy Bundle / Create Bundle. The copy bundle file is a compressed version of your entire project and saved as one file with the extension acb (Atlas Copy Bundle). If needed you can unpack the bundle to restore your project: Tools / Copy Bundle / Install Bundle.
Codes
Codes are used as classification devices at different levels of abstraction in order to create sets of related information units for the purpose of comparison.
Coding Analyzer
The Coding Analyzer is helpful after merging HUs. It tests for redundant codes, i.e., codes used more than once for data segments that overlap or are embedded in one another.
Comments
Everything can be commented in ATLAS.ti! Comments are not "first class objects" themselves. They do not have a name, they do not have their own list window, nor do they have any property other
APPENDIX
144
than the text they contain. However, they are an important attribute of the objects that they attached to. Writing is one of the main activities when working with ATLAS.ti. Although the program can't tell you what to write, it does support you in attaching your writings to the most appropriate location. All objects have a "slot" that can be filled with a comment. The first comment you write is typically a comment for the HU, which appears in the HU browser and in output created with the HTML and XML generator. But even links and relations can be commented.
Cooccurence Explorer
The Cooccurence Explorer is more exploratory than the Query Tool. Rather than determining the codes yourself, you can ask ATLAS.ti which codes happen to cooccur in the margin area. The output can be viewed in form or a tree view or a table view. The table view provides frequency of cooccurence and a correlation coefficient if appropriate. At the same time, you can access the data behind the numbers.
Copy Bundle
The Copy Bundle function serves a dual purpose: Portability (to migrate a project to another location), and data security (a powerful backup and restore device). You should exclusively use Copy Bundle to port your projects between computers. To preserve the consistency of a project in regard to referenced files, refrain from copying projects manually using Windows file procedures. ALWAYS use Copy Bundle instead! By inspecting the HU, it finds and collects all files that make up the project, the data files used as PDs, and all associated auxiliary files. It checks the accessibility of the data sources and provides feedback in problematic situations. From all the project files it compiles a single compressed file which you can then easily safely store or transfer to another computer. On the target computer, installing a bundle distributes the HU, the data source files, and all associated files to appropriate location(s).
Families
Families are a way to form clusters of PDs, codes, and memos for easier handling of groups of codes, memos, and PDs. Primary Document families can be regarded as attributes or variables. For more detail, see the chapter Family Life in the full manual. Families can be combined using logical operators similar to codes and Super Codes.
APPENDIX
145
Family Manager
Codes, memos, and PDs can be grouped within "families". The Family Manager is the tool with which you create, modify, and edit them.
Hermeneutic Unit
The Hermeneutic Unit (HU), provides the data structure for each project in ATLAS.ti. Everything that is relevant to a particular project (e.g., a research topic) is part of the HU and resides in the electronic environment! For instance, the Primary Documents representing the data sources, the quotations, the codes, the conceptual linkages (families, networks), and the memos, etc., are all part of one HU. The lowest level of an HU contains the Primary Documents, followed closely by the "quotations" as selections of the Primary Documents. On the next level, codes refer to quotations. Memos - you meet them everywhere.
HTML Generator
Publishing online or just creating a printout from a project is available with the HTML generator. The results can be viewed platform independent with any Web browser.
Link Managers
The Link Managers provide an overview of all code-code links and of all quotation-quotation links you have created.
APPENDIX
146
Margin Area
To the right hand side of the PD area is the margin area. Bars are used to indicate quotations. Attached codes, memos, and hyperlinks are displayed next to the brackets. The margin area is fully interactive: The displayed objects are sensitive to mouse clicks and have their own context menus. In addition, a variety of dragand-drop options are possible. A general context menu can be opened inside the margin area by moving the mouse pointer onto the background between the objects. It allows you to set display properties of the objects.
Memos
Memos capture your thoughts regarding the text and are an important device for creating theory. A "memo" is similar to a code, but usually contains longer passages of text. A memo may "stand alone" or it may refer to quotations, codes, and other memos. They can be grouped according to types (method, theoretical, descriptive, etc.), which is helpful in organizing and sorting them. Memos may also be included as the objects of analysis by assigning them as PDs.
Network Editor
The Network Editor displays and offers all editing capability to construct and refine semantic networks. In addition, it allows the visual creation and traversal of hypertext structures.
Network Views
Network Views are a bit more sophisticated than families. They allow you to conceptualize the structure by connecting sets of similar elements together in a visual diagram. With the aid of Network Views you can express relationships between codes, quotations, and memos. PDs, families and even Network Views can also be nodes in a network view.
APPENDIX
147
Relations are link prototypes used to create a link between two codes or between two quotations. An example is the ... is a... (ISA) relation, which is frequently used to link concepts of different abstraction level (e.g., DOG <isa> MAMMAL).
Object Crawler
With the Object Crawler, you can search all of the parts of your project within ATLAS.ti that contain text. Searches are not restricted to just textual PDs: codes, memos, quotations, all families, code-code links, hyper-links, and the HU can be searched. In addition, the scope of the search can be limited to certain fields.
Object Explorer
The Object Explorer displays all the elements of an HU in a strictly hierarchical manner, even if the structures are non-hierarchical, or even cyclic.
Object Managers
Each main object list (primary document, quotation, code, memo) can be displayed and managed in a separate window called Object Manager. The Managers offer a variety of additional functions and are probably the most frequently used tools besides the HU Editor.
Primary Documents
Primary Documents (PDs) represent data sources, i.e., the text, image, audio, video or geographic materials that you wish to interpret. The content of PDs is usually stored in data files on your computer. Within ATLAS.ti, a primary document is created by assigning a data file to an HU. You can, however, also create internal text documents and assign memos as a primary documents. When using data files as sources for your PDs, then you need to make sure that access to these sources is provided. This means the source file has to remain at the referenced location and should not be renamed; otherwise the HU cannot find it. To avoid jeopardizing the integrity of your coding, data sources should never be changed outside ATLAS.ti. In case you need to make changes to your data sources, ATLAS.ti offers an edit mode. You can assign as many documents as needed for a given HU. See Assigning and Handling Primary Documents in the full manual for further details.
APPENDIX
148
family called female. All male interviewees into a PD family called male. You can do the same for different professions, marital status, education levels, etc. The classification is a 0 / 1 classification. 1 means the document is part of a particular group; 0 means it is not part of the group. Following a special syntax, you can also create PD families that can be imported or exported as variables with two or more characteristics based of an Excel table like Gender::male and Gender::female. PD Families can be later used to restrict code based searches like: Show me all data segments coded with attitude towards the environment but only for females who live in London as compared to females who live in the country side. You can also use PD families as filer to reduce other type of output like a frequency count for codes across a particular group of documents.
Project Transfer
In addition to using the the Copy Bundle function as backup, you can also use it for project transfer: TOOLS / COPY BUNDLE / CREATE BUNDLE. To unpack a project bundle at the target location, select: TOOLS / COPY BUNDLE / INSTALL BUNDLE.
Query Tool
For more complex search requests, the Query Tool is at your disposal. Here you can formulate search requests that are based on combinations of codes using one or a combination of 14 different operators (Boolean, semantic and proximity operators)
Quotations
A quotation is a segment from a PD that is interesting or important to the user. In textual documents, a quotation is an arbitrary sequence of characters ranging from a single character, a word, a sentence, or a paragraph up to the entire data file. Free quotations resemble passages "scribbled" in the margin of a book. Usually quotations are created by the researcher. However, if repetitive words or phrases are found in the text, the Auto-Coding feature can be used to automatically segment these quotations and assign a code to them. When a quotation is created, ATLAS.ti automatically assigns an identifier to it. This identifier is built from the index of the primary text to which it belongs and the first 30 letters (note that a different length can be set via Preferences) of the text segment, e.g., "1:21 Therefore a more efficient fil....". The identifier is displayed in list windows and printouts. For graphic, audio, and video segments, the original file name of the PD is chosen as an identifier. Although the creation of quotations is almost always part of a broader task like coding or writing memos, "free" quotations can be created that
APPENDIX
149
indicate interesting parts in the primary data for which a meaningful classification has not yet been found.
Relation Editor
When the built-in relations that are used to connect objects in Network Views are not sufficient, you can edit them or create new ones using the Relation Editor.
Super Code
A Super Code differs from a standard code: A standard code is directly linked with the quotations to which it is associated, while a Super Code is a stored query. Thus it provides an answer to a question (in the best case) that typically consists of several combined codes. See the chapter on the Query Tool in the full manual.
Text Editor
Text editors can be opened to edit comments, to create and edit memos, and to view, format, and print results. All editors are capable of displaying and editing rich text format (RTF). They offer WordPad-equivalent WYSIWYG convenience and editing features.
Variables
See Primary Document Families.
Word Cruncher
The Word Cruncher counts all words in textual PDs. The count can be limited to one PD only. To clean up the count, a stop list can be defined to exclude special symbols or words like and, or, the, etc. The result can be displayed in an Excel table.
APPENDIX
150
XML Converter
The XML Converter lists HUs saved in XML format and applies so-called "style sheets" (miniature programs written in XSLT) to the XML files. The user may edit the supplied style sheets or add new ones.
XML Generator
The XML generator exports all information contained in an HU in XML format. This universal, open data format allows for a wide variety of possibilities for display, processing, and even integration of your data with external applications.
User Administration
Manage the ATLAS.ti user database through the user administration tool. This is a prerequisite for collaborative work, but is also useful to individual users through personalizing the login or protecting an HU with a password. See User Management in the full manual.